Reproducible Research Project 2

Storm Events with Most Effects

on Population Health (Fatalities, Injuries),

and Economic Consequences (Crop and Property)

Analysis of the NOAA Storm Database (1950-2011), United States

8 Februari 2015

Synopsis

This analysis report explores the U.S. National Oceanic and Atmospheric Administration's (NOAA) Storm Database and answering some basic questions about severe weather events through some data processing with results presented in visual plots and tables.
The analysis trying to resolve questions:
1. Which types of events (as indicated in the EVTYPE variable) are most harmful with respect to population health across the United States?
2. Which types of events have the greatest economic consequences across the United States?

Introduction

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 damages, and preventing such outcomes to the extent possible is a key concern.

The NOAA 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.

Database

Data repository:
Storm Data [47Mb]
Documentation:
National Weather Service Storm Data Documentation
National Climatic Data Center Storm Events FAQ


Data Processing

library(R.utils)   
library(ggplot2)   
library(knitr)   
library(rapport)   
library(reshape2)   
require(gridExtra)  
opts_chunk$set(cache=TRUE)   

Download database

db_url <- "https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2"   
db_zip <- "StormData.csv.bz2"   
db_csv <- "StormData.csv"   

if (!file.exists(db_zip)) {    
    download.file(db_url, db_zip, method="curl")    
}    

Extract database

if (!file.exists(db_csv)){    
    bunzip2(db_zip, db_csv, remove=FALSE)   
}   

Read as data frame

data <- read.csv(db_csv, sep=",", header=TRUE)   

Explore data by colnames(data), head(data), or str(data.

str(data)
## 'data.frame':    902297 obs. of  37 variables:
##  $ STATE__   : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ BGN_DATE  : Factor w/ 16335 levels "1/1/1966 0:00:00",..: 6523 6523 4242 11116 2224 2224 2260 383 3980 3980 ...
##  $ BGN_TIME  : Factor w/ 3608 levels "00:00:00 AM",..: 272 287 2705 1683 2584 3186 242 1683 3186 3186 ...
##  $ TIME_ZONE : Factor w/ 22 levels "ADT","AKS","AST",..: 7 7 7 7 7 7 7 7 7 7 ...
##  $ COUNTY    : num  97 3 57 89 43 77 9 123 125 57 ...
##  $ COUNTYNAME: Factor w/ 29601 levels "","5NM E OF MACKINAC BRIDGE TO PRESQUE ISLE LT MI",..: 13513 1873 4598 10592 4372 10094 1973 23873 24418 4598 ...
##  $ STATE     : Factor w/ 72 levels "AK","AL","AM",..: 2 2 2 2 2 2 2 2 2 2 ...
##  $ EVTYPE    : Factor w/ 985 levels "   HIGH SURF ADVISORY",..: 834 834 834 834 834 834 834 834 834 834 ...
##  $ BGN_RANGE : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ BGN_AZI   : Factor w/ 35 levels "","  N"," NW",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ BGN_LOCATI: Factor w/ 54429 levels "","- 1 N Albion",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ END_DATE  : Factor w/ 6663 levels "","1/1/1993 0:00:00",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ END_TIME  : Factor w/ 3647 levels ""," 0900CST",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ COUNTY_END: num  0 0 0 0 0 0 0 0 0 0 ...
##  $ COUNTYENDN: logi  NA NA NA NA NA NA ...
##  $ END_RANGE : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ END_AZI   : Factor w/ 24 levels "","E","ENE","ESE",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ END_LOCATI: Factor w/ 34506 levels "","- .5 NNW",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ LENGTH    : num  14 2 0.1 0 0 1.5 1.5 0 3.3 2.3 ...
##  $ WIDTH     : num  100 150 123 100 150 177 33 33 100 100 ...
##  $ F         : int  3 2 2 2 2 2 2 1 3 3 ...
##  $ MAG       : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ FATALITIES: num  0 0 0 0 0 0 0 0 1 0 ...
##  $ INJURIES  : num  15 0 2 2 2 6 1 0 14 0 ...
##  $ PROPDMG   : num  25 2.5 25 2.5 2.5 2.5 2.5 2.5 25 25 ...
##  $ PROPDMGEXP: Factor w/ 19 levels "","-","?","+",..: 17 17 17 17 17 17 17 17 17 17 ...
##  $ CROPDMG   : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ CROPDMGEXP: Factor w/ 9 levels "","?","0","2",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ WFO       : Factor w/ 542 levels ""," CI","$AC",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ STATEOFFIC: Factor w/ 250 levels "","ALABAMA, Central",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ ZONENAMES : Factor w/ 25112 levels "","                                                                                                                               "| __truncated__,..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ LATITUDE  : num  3040 3042 3340 3458 3412 ...
##  $ LONGITUDE : num  8812 8755 8742 8626 8642 ...
##  $ LATITUDE_E: num  3051 0 0 0 0 ...
##  $ LONGITUDE_: num  8806 0 0 0 0 ...
##  $ REMARKS   : Factor w/ 436774 levels "","-2 at Deer Park\n",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ REFNUM    : num  1 2 3 4 5 6 7 8 9 10 ...

Keep only columns related to event type, cumulative fatalities, injuries, and damages.
EXP is exponent (unit of measure), i.e. use unique(data$CROPDMGEXP) to explore possible values.
See below for exponent explanation.

cols <- c("EVTYPE", "FATALITIES", "INJURIES", "CROPDMG", "CROPDMGEXP", "PROPDMG", "PROPDMGEXP")   
data <- subset(data, select = cols)   

# trim leading and trailing spaces of EVTYPE character   
data$EVTYPE <- trim.space(data$EVTYPE)   
# all upper case to ease data cleaning   
data$EVTYPE <- toupper(data$EVTYPE)   

dim(data)  
## [1] 902297      8
head(data)  
##   X  EVTYPE FATALITIES INJURIES CROPDMG CROPDMGEXP PROPDMG PROPDMGEXP
## 1 1 TORNADO          0       15       0               25.0          K
## 2 2 TORNADO          0        0       0                2.5          K
## 3 3 TORNADO          0        2       0               25.0          K
## 4 4 TORNADO          0        2       0                2.5          K
## 5 5 TORNADO          0        2       0                2.5          K
## 6 6 TORNADO          0        6       0                2.5          K

Some EVTYPE values has abbreviations,

## substitute abbreviations   
data$EVTYPE <- gsub("CSTL", "COASTAL", data$EVTYPE)  
data$EVTYPE <- gsub("FLDG", "FLOOD", data$EVTYPE)
data$EVTYPE <- gsub("HVY", "HEAVY", data$EVTYPE)  
data$EVTYPE <- gsub("SML", "SMALL", data$EVTYPE)
data$EVTYPE <- gsub("STRM", "STREAM", data$EVTYPE)
data$EVTYPE <- gsub("TSTMW", "THUNDERSTORM WIND", data$EVTYPE)  
data$EVTYPE <- gsub("TSTM", "THUNDERSTORM", data$EVTYPE)  
data$EVTYPE <- gsub("VOG", "FOG", data$EVTYPE)  
data$EVTYPE <- gsub("WND", "WIND", data$EVTYPE)  

Exponent values of economic damages, CROPDMGEXP and PROPDMGEXP:

How to resolve this symbol, was discussed in the course's forum.
Hints are provided by David Hood (Data Science Specialization's CTA), as well as by Eddie Song (class fellow in course). Credit goes to them.

This is analysis compilation of this exponent values in my RPubs repo, or in Markdown or PDF file, on how to handle these exponent values.

Set CROPDMGEXP and PROPDMGEXP values to multiplier number

## 0..8  
data$CROPDMGEXP <- gsub("[[:digit:]]", "10", data$CROPDMGEXP)    
data$PROPDMGEXP <- gsub("[[:digit:]]", "10", data$PROPDMGEXP)    
## +    
data$CROPDMGEXP <- gsub("\\+", "1", data$CROPDMGEXP)    
data$PROPDMGEXP <- gsub("\\+", "1", data$PROPDMGEXP)    
## -,?  
data$CROPDMGEXP <- gsub("[-\\?]", "0", data$CROPDMGEXP)    
data$PROPDMGEXP <- gsub("[-\\?]", "0", data$PROPDMGEXP)    
## H,h, K,k, M,m, B,b   
data$CROPDMGEXP <- gsub("[Hh]", "100", data$CROPDMGEXP)    
data$CROPDMGEXP <- gsub("[Kk]", "1000", data$CROPDMGEXP)    
data$CROPDMGEXP <- gsub("[Mm]", "1000000", data$CROPDMGEXP)    
data$CROPDMGEXP <- gsub("[Bb]", "1000000000", data$CROPDMGEXP)    
data$PROPDMGEXP <- gsub("[Hh]", "100", data$PROPDMGEXP)    
data$PROPDMGEXP <- gsub("[Kk]", "1000", data$PROPDMGEXP)    
data$PROPDMGEXP <- gsub("[Mm]", "1000000", data$PROPDMGEXP)    
data$PROPDMGEXP <- gsub("[Bb]", "1000000000", data$PROPDMGEXP)    

## clean exponent data   
data$CROPDMGEXP[data$CROPDMGEXP == ""] <- 0 # empty character as 0    
data$PROPDMGEXP[data$PROPDMGEXP == ""] <- 0   
data$CROPDMGEXP <- as.numeric(data$CROPDMGEXP) # convert to numeric class   
data$PROPDMGEXP <- as.numeric(data$PROPDMGEXP)   

## calculate $ value = (number * exponent-number)   
data <- mutate(data, NEWCROPDMG = CROPDMG * CROPDMGEXP)   
data <- mutate(data, NEWPROPDMG = PROPDMG * PROPDMGEXP)   

Results

1. Types of events which are most harmful with respect to population health:

## summarize population health, injuries, fatalities   
total <- aggregate(cbind(INJURIES, FATALITIES) ~ EVTYPE, data, FUN=sum, na.rm=TRUE)   
## sort and filter top only    
total <- arrange(total, desc(FATALITIES + INJURIES))[1:12,]   
## convert wide to long format  
totalLong <- melt(total, id.vars="EVTYPE")   

Bar plot, total stacked:

xaxis <- reorder(totalLong$EVTYPE, -(totalLong$value))    
gp <- ggplot(aes(x=xaxis, y=value, fill=variable), data=totalLong) +  
  geom_bar(stat="identity", position="stack", colour="black") +  
  scale_y_continuous(breaks=seq(from=0, to=1e5, by=5000)) +  
  labs(x="Event Type", y="Number of Population Affected",  
       title="Most Harmful Events to Population Health in the United States \n(Total Stacked, Injuries and Fatalities)") +  
  theme(axis.text.x=element_text(angle=40, hjust=1))  
print(gp)    

plot of chunk unnamed-chunk-11

The plots and table show us that tornado, thunderstorm wind and excessive heat causes most harms to the population health.

Table of most harmful events to population health:

table_total <- mutate(total, Total = INJURIES + FATALITIES)    
table_total <- format(table_total, big.mark=",", decimal.mark=".")    
kable(table_total, format="html", col.names=c("Event Type", "Injuries", "Fatalities", "Total"), 
      align=c("l", "r", "r", "r"))   
Event Type Injuries Fatalities Total
TORNADO 91,346 5,633 96,979
THUNDERSTORM WIND 8,445 637 9,082
EXCESSIVE HEAT 6,525 1,903 8,428
FLOOD 6,789 470 7,259
LIGHTNING 5,230 816 6,046
HEAT 2,100 937 3,037
FLASH FLOOD 1,777 978 2,755
ICE STORM 1,975 89 2,064
WINTER STORM 1,321 206 1,527
HIGH WIND 1,137 248 1,385
HAIL 1,361 15 1,376
HURRICANE/TYPHOON 1,275 64 1,339

2. Types of events which have greatest economic consequences

## summarize, economic damages, crop and property   
damage <- aggregate(cbind(NEWPROPDMG, NEWCROPDMG) ~ EVTYPE, data, FUN=sum, na.rm=TRUE)   
## sort and filter top only    
damage <- arrange(damage, desc(NEWPROPDMG + NEWCROPDMG))[1:12,]   
## convert wide to long format  
damageLong <- melt(damage, id.vars="EVTYPE")   

Bar plot, total stacked, economic damages

xaxis <- reorder(damageLong$EVTYPE, -(damageLong$value))    
gp <- ggplot(aes(x=xaxis, y=value/1e9, fill=variable), data=damageLong) +  
  geom_bar(stat="identity", position="stack", colour="black") +  
  scale_y_continuous(breaks=seq(from=0, to=150, by=10)) +  
  labs(x="Event Type", y="Economic damages ($ billion)",  
       title="Greatest Economic Consequences in the United States \n(Total Stacked, Property and Crop)") +  
  theme(axis.text.x=element_text(angle=40, hjust=1))  
print(gp)    

plot of chunk unnamed-chunk-15

The plots and table show us that, flood, hurricane / typhoon and tornado causes most economic damages.

Table of events with greatest economic consequences:

table_damage <- mutate(damage, Total = NEWPROPDMG + NEWCROPDMG)    
table_damage <- format(table_damage, big.mark=",", decimal.mark=".")    
kable(table_damage, format="html", col.names=c("Event Type", "Property Damages", "Crop Damages", "Total"), 
      align=c("l", "r", "r", "r"))   
Event Type Property Damages Crop Damages Total
FLOOD 144,657,709,800 5,661,968,450 150,319,678,250
HURRICANE/TYPHOON 69,305,840,000 2,607,872,800 71,913,712,800
TORNADO 56,937,162,897 414,954,710 57,352,117,607
STORM SURGE 43,323,536,000 5,000 43,323,541,000
HAIL 15,732,269,877 3,025,954,650 18,758,224,527
FLASH FLOOD 16,140,865,011 1,421,317,100 17,562,182,111
DROUGHT 1,046,106,000 13,972,566,000 15,018,672,000
HURRICANE 11,868,319,010 2,741,910,000 14,610,229,010
RIVER FLOOD 5,118,945,500 5,029,459,000 10,148,404,500
ICE STORM 3,944,928,310 5,022,113,500 8,967,041,810
THUNDERSTORM WIND 7,976,231,550 968,850,400 8,945,081,950
TROPICAL STORM 7,703,890,550 678,346,000 8,382,236,550