NOAA Storm Database Exploratory Analysis
Synopsis
This work is intended to answer some proposed questions, by analysing the US National Oceanic and Atmospheric Administration’s (NOAA) Storm database. Complete information about database can be found at Storm Data Documentation. The goal of such database is to map characteristics of storms and other weather phenomena.
More specific, those main questions are:
Across the United States, 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?
Therefore, report explains how data are processed and at last, show results with some variety of formats, like figures and tables, for example. Main objective here is to exercise exploratory analysis data flow, Starting by collect data, feature selection and some adjustments, and finally get some first insights from data by graphical analysis.
Data processing
As input of analysis, I used the NOAA Storm Database. This database looks like these, in raw data format:
data <- read.csv('repdata-data-StormData.csv')
head(data)## STATE__ BGN_DATE BGN_TIME TIME_ZONE COUNTY COUNTYNAME STATE
## 1 1 4/18/1950 0:00:00 0130 CST 97 MOBILE AL
## 2 1 4/18/1950 0:00:00 0145 CST 3 BALDWIN AL
## 3 1 2/20/1951 0:00:00 1600 CST 57 FAYETTE AL
## 4 1 6/8/1951 0:00:00 0900 CST 89 MADISON AL
## 5 1 11/15/1951 0:00:00 1500 CST 43 CULLMAN AL
## 6 1 11/15/1951 0:00:00 2000 CST 77 LAUDERDALE AL
## EVTYPE BGN_RANGE BGN_AZI BGN_LOCATI END_DATE END_TIME COUNTY_END
## 1 TORNADO 0 0
## 2 TORNADO 0 0
## 3 TORNADO 0 0
## 4 TORNADO 0 0
## 5 TORNADO 0 0
## 6 TORNADO 0 0
## COUNTYENDN END_RANGE END_AZI END_LOCATI LENGTH WIDTH F MAG FATALITIES
## 1 NA 0 14.0 100 3 0 0
## 2 NA 0 2.0 150 2 0 0
## 3 NA 0 0.1 123 2 0 0
## 4 NA 0 0.0 100 2 0 0
## 5 NA 0 0.0 150 2 0 0
## 6 NA 0 1.5 177 2 0 0
## INJURIES PROPDMG PROPDMGEXP CROPDMG CROPDMGEXP WFO STATEOFFIC ZONENAMES
## 1 15 25.0 K 0
## 2 0 2.5 K 0
## 3 2 25.0 K 0
## 4 2 2.5 K 0
## 5 2 2.5 K 0
## 6 6 2.5 K 0
## LATITUDE LONGITUDE LATITUDE_E LONGITUDE_ REMARKS REFNUM
## 1 3040 8812 3051 8806 1
## 2 3042 8755 0 0 2
## 3 3340 8742 0 0 3
## 4 3458 8626 0 0 4
## 5 3412 8642 0 0 5
## 6 3450 8748 0 0 6Among all those variables below, our concern is about EVTYPE, which means type of events related to population health.
names(data)## [1] "STATE__" "BGN_DATE" "BGN_TIME" "TIME_ZONE" "COUNTY"
## [6] "COUNTYNAME" "STATE" "EVTYPE" "BGN_RANGE" "BGN_AZI"
## [11] "BGN_LOCATI" "END_DATE" "END_TIME" "COUNTY_END" "COUNTYENDN"
## [16] "END_RANGE" "END_AZI" "END_LOCATI" "LENGTH" "WIDTH"
## [21] "F" "MAG" "FATALITIES" "INJURIES" "PROPDMG"
## [26] "PROPDMGEXP" "CROPDMG" "CROPDMGEXP" "WFO" "STATEOFFIC"
## [31] "ZONENAMES" "LATITUDE" "LONGITUDE" "LATITUDE_E" "LONGITUDE_"
## [36] "REMARKS" "REFNUM"Looking closer to EVTYPE, we can see some examples of these events.
head(unique(data$EVTYPE))## [1] TORNADO TSTM WIND HAIL
## [4] FREEZING RAIN SNOW ICE STORM/FLASH FLOOD
## 985 Levels: ? ABNORMALLY DRY ABNORMALLY WET ... WNDSo, to begin this analysis. We have chosen four variables: STATE, EVTYPE, FATALITIES, INJURIES and PROPDMG (property damage)
library(data.table)
harmfulEvent <- data.table(data$STATE, data$EVTYPE, data$FATALITIES, data$INJURIES, data$PROPDMG)
setnames(harmfulEvent, c('state','evtype','fatalities','injuries', 'propdmg'))
head(harmfulEvent)## state evtype fatalities injuries propdmg
## 1: AL TORNADO 0 15 25.0
## 2: AL TORNADO 0 0 2.5
## 3: AL TORNADO 0 2 25.0
## 4: AL TORNADO 0 2 2.5
## 5: AL TORNADO 0 2 2.5
## 6: AL TORNADO 0 6 2.5At last, we can make some aggregations in order to discover how many fatalities each state most suffer by event.
library(plyr)
ag <- harmfulEvent[, sum(fatalities), by = c('evtype','state')]
max_harm_state <- ddply(ag,~state,function(x){x[which.max(x$V1),]})
sortedHarm <- max_harm_state[order(-max_harm_state$V1),]
head(sortedHarm)## evtype state V1
## 20 HEAT IL 653
## 2 TORNADO AL 617
## 63 TORNADO TX 538
## 38 TORNADO MS 450
## 37 TORNADO MO 388
## 5 TORNADO AR 379Notice that state of Illinois suffers with heat, followed by Alabama, which tornado is main cause of fatalities. Similarly, same analysis can be done with injuries variable:
ag <- harmfulEvent[, sum(injuries), by = c('evtype','state')]
max_injuries_state <- ddply(ag,~state,function(x){x[which.max(x$V1),]})
sortedInjuries <- max_injuries_state[order(-max_injuries_state$V1),]
head(sortedInjuries)## evtype state V1
## 63 TORNADO TX 8207
## 2 TORNADO AL 7929
## 38 TORNADO MS 6244
## 5 TORNADO AR 5116
## 49 TORNADO OK 4829
## 62 TORNADO TN 4748Summary above clearly states that tornadoes are high related to injuries. Finally, we do same analysis for property damage (propdmg variable), which give us some clue about economic consequences:
ag <- harmfulEvent[, sum(propdmg), by = c('evtype','state')]
max_propdmg_state <- ddply(ag,~state,function(x){x[which.max(x$V1),]})
sortedPropdmg <- max_propdmg_state[order(-max_propdmg_state$V1),]
head(sortedPropdmg)## evtype state V1
## 63 TORNADO TX 283097.2
## 38 TORNADO MS 187840.9
## 2 TORNADO AL 167816.2
## 49 TORNADO OK 165167.9
## 13 TORNADO FL 159752.6
## 18 TORNADO IA 152142.8Again, tornadoes are main cause of property damage.
Results
As results, we can show histograms of previous section analysis. All plots just contain top 20 fatalities, injuries and property damages, respectivelly, grouped by state
library(ggplot2)
topFatalities <- sortedHarm[1:20,]
p <- qplot(x = topFatalities$state, y = topFatalities$V1, topFatalities,
color = topFatalities$evtype, geom = c('point','smooth'), method = "lm",
xlab = "state", ylab = "num of fatalities", main = 'Num fatalities x state')
print(p, width = 100)
Injuries plot by state is very similar to above fatalities histogram:
library(ggplot2)
topInjuries <- sortedInjuries[1:20,]
p <- qplot(x = topInjuries$state, y = topInjuries$V1, topInjuries,
color = topInjuries$evtype, geom = c('point','smooth'), method = "lm",
xlab = "state", ylab = "num of injuries", main = 'Num injuries x state')
print(p, width = 50)
Last plot shows relation between property damage and event type:
library(ggplot2)
topDamaged <- sortedPropdmg[1:20,]
p <- qplot(x = topDamaged$state, y = topDamaged$V1, topDamaged,
color = topInjuries$evtype, geom = c('point','smooth'), method = "lm",
xlab = "state", ylab = "property damages", main = 'Property damages x state')
print(p, width = 50)