Exploring the U.S. National Ocenaic and Atmopheric Administrations (NOAA) storm database - health and economic consequence of severe wheather

Synopsis

As part of the second project in the Coursera course Reproducible Research we are asked to analyse economic and health consequencess of severe weather event during yhe years 1950 - 2011 using the NOAA database. The NOAA database tracks storms and weather events in the US. We are asked to answer the questions

Q1: Across the United States, which types of events are most harmful with respect to population health?
Q2; Across the United States, which types of events have the greatest economic consequences?

The analysis shows, that tornados are the most harmful events in respect to population health and floods have the greatest economic consequences.

Data Processing

1 Loading packages The packages must be installed prior to running the code

library(ggplot2)
library(dplyr)
library(plyr)
library(Hmisc)
library(flextable)

2 Reading the data Assuming data has been downloaded to working directory. The storm data can found at:
NoAA data

stormfile<-"repdata_data_StormData.csv.bz2"
rawdata <- read.csv(file = stormfile, header=TRUE, sep=",")

3 Transforming data and subsetting
According to NOAA, the data recording start from Jan. 1950. At that time, only one event type was recorded - tornado. More events gradually were added, and only from Jan 1996 all event types have been recorded. Since our objective is comparing the effects of different weather events, we need only to include events that started not earlier than Jan 1996.

# Reading data
rawdata$BGN_DATE <- strptime(rawdata$BGN_DATE, "%m/%d/%Y %H:%M:%S")
maindata <- subset(rawdata, BGN_DATE > "1995-12-31")
rm(rawdata)

4 Selecting variables
Based on exploration of the raw data with the functions ‘str’, ‘names’, ‘dim’, ‘head’, ‘range’ I conclude that there are 7 variables we are interested in regarding the two questions.Namely: EVTYPE, FATALITIES, INJURIES, PROPDMG, PROPDMGEXP, CROPDMG, CROPDMGEXP.

EVTYPE: Type of event
FATALITIES: Number of fatalities
INJURIES: Number of injuries
PROPDMG: Size of property damage
PROPDMGEXP: The exponent values for PROPDMG
CROPDMG: Size of crop damage
CROPDMGEXP: The exponent values for CROPDMG

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

5 There are many unique event types in EVTYPE column. We simplify by capitalizing all letters in EVTYPE column. Further, I subset only non-zero data regarding our target numbers

maindata$EVTYPE <- toupper(maindata$EVTYPE)

maindata <- maindata[maindata$FATALITIES !=0 | 
                       maindata$INJURIES !=0 | 
                       maindata$PROPDMG !=0 | 
                       maindata$CROPDMG !=0, ]

Results

We are now ready to answer the question:

Q1: Across the United States, which types of events (as indicated in the EVTYPE variable) are most harmful with respect to population health?

#Sum fatalities ad injuries by Event Type
fatalities <- aggregate(FATALITIES ~ EVTYPE, data=maindata, sum)
injuries <- aggregate(INJURIES ~ EVTYPE, data=maindata, sum)

#Arrange in descending order by Event Type by number of fatalities or injuries - extract top ten
fatalities <- arrange(fatalities,desc(FATALITIES),EVTYPE)[1:10,]
injuries <- arrange(injuries,desc(INJURIES),EVTYPE)[1:10,]

Fatalities:

fatalities

##            EVTYPE FATALITIES
## 1  EXCESSIVE HEAT       1797
## 2         TORNADO       1511
## 3     FLASH FLOOD        887
## 4       LIGHTNING        651
## 5           FLOOD        414
## 6     RIP CURRENT        340
## 7       TSTM WIND        241
## 8            HEAT        237
## 9       HIGH WIND        235
## 10      AVALANCHE        223

Injuries:

injuries

##               EVTYPE INJURIES
## 1            TORNADO    20667
## 2              FLOOD     6758
## 3     EXCESSIVE HEAT     6391
## 4          LIGHTNING     4141
## 5          TSTM WIND     3629
## 6        FLASH FLOOD     1674
## 7  THUNDERSTORM WIND     1400
## 8       WINTER STORM     1292
## 9  HURRICANE/TYPHOON     1275
## 10              HEAT     1222

From the output it is evident that tornados are the most harmful weather event with respect to health looking across both fatalities and injuries.

Supporting charts:

# Fatalities per event type
ggplot(fatalities, aes(x = EVTYPE, y = FATALITIES)) + 
      geom_bar(stat = "identity", fill = "red", width = NULL) + 
      theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
      xlab("Event Type") + ylab("Fatalities")

# Injuries per event type
ggplot(injuries, aes(x = EVTYPE, y = INJURIES)) + 
      geom_bar(stat = "identity", fill = "red", width = NULL) + 
      theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
      xlab("Event Type") + ylab("Injuries")

Lets turn to the next assignment:

Q2: Across the United States, which types of events have the greatest economic consequences?

Wee need to convert exponents to real numbers.

maindata$PROPDMGEXP <- gsub("[Hh]", "2", maindata$PROPDMGEXP)
maindata$PROPDMGEXP <- gsub("[Kk]", "3", maindata$PROPDMGEXP)
maindata$PROPDMGEXP <- gsub("[Mm]", "6", maindata$PROPDMGEXP)
maindata$PROPDMGEXP <- gsub("[Bb]", "9", maindata$PROPDMGEXP)
maindata$PROPDMGEXP <- gsub("\\+", "1", maindata$PROPDMGEXP)
maindata$PROPDMGEXP <- gsub("\\?|\\-|\\ ", "0",  maindata$PROPDMGEXP)
maindata$PROPDMGEXP <- as.numeric(maindata$PROPDMGEXP)

maindata$CROPDMGEXP <- gsub("[Hh]", "2", maindata$CROPDMGEXP)
maindata$CROPDMGEXP <- gsub("[Kk]", "3", maindata$CROPDMGEXP)
maindata$CROPDMGEXP <- gsub("[Mm]", "6", maindata$CROPDMGEXP)
maindata$CROPDMGEXP <- gsub("[Bb]", "9", maindata$CROPDMGEXP)
maindata$CROPDMGEXP <- gsub("\\+", "1", maindata$CROPDMGEXP)
maindata$CROPDMGEXP <- gsub("\\-|\\?|\\ ", "0", maindata$CROPDMGEXP)
maindata$CROPDMGEXP <- as.numeric(maindata$CROPDMGEXP)

maindata$PROPDMGEXP[is.na(maindata$PROPDMGEXP)] <- 0
maindata$CROPDMGEXP[is.na(maindata$CROPDMGEXP)] <- 0

maindata <- mutate(maindata, 
                    PROPDMGTOTAL = PROPDMG * (10 ^ PROPDMGEXP), 
                    CROPDMGTOTAL = CROPDMG * (10 ^ CROPDMGEXP))

# Summing economic consequencess
Economic_data <- aggregate(cbind(PROPDMGTOTAL, CROPDMGTOTAL) ~ EVTYPE, data = maindata, FUN=sum)
Economic_data$ECONOMIC_LOSS <- Economic_data$PROPDMGTOTAL + Economic_data$CROPDMGTOTAL
Economic_data <- Economic_data[order(Economic_data$ECONOMIC_LOSS, decreasing = TRUE), ]
worsteconomicevents <- Economic_data[1:10,c(1,4)]
worsteconomicevents

##                EVTYPE ECONOMIC_LOSS
## 48              FLOOD  148919611950
## 88  HURRICANE/TYPHOON   71913712800
## 141       STORM SURGE   43193541000
## 149           TORNADO   24900370720
## 66               HAIL   17071172870
## 46        FLASH FLOOD   16557105610
## 86          HURRICANE   14554229010
## 32            DROUGHT   14413667000
## 152    TROPICAL STORM    8320186550
## 83          HIGH WIND    5881421660

# Loss per event type
ggplot(worsteconomicevents, aes(x = EVTYPE, y = ECONOMIC_LOSS)) + 
  geom_bar(stat = "identity", fill = "blue") + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) + 
  xlab("Event Type") + ylab("Total Prop & Crop Damages (USD)") +
ggtitle("Total economic loss in the US in the period 1996 - 2011 by weather event")