Reproducible Research Project 2
Durai
October 23, 2015
An Analysis of the Impact of Storms in the US
Synopsis
In this analysis, we import data from the NOAA and aggregate by event (storm) type to show the most signifcant:
Population health impacts Economic impacts
TO do this, we load the bzip file into a dataframe, then aggregate the “EVTYPE” by fields “FATALITIES” and “INJURIES” for population health, and by “CROPDMG” and “PROPDMG” for economic impacts.
We’re interested in the most impactful events, as such we use the top 5% of the data. While it may seem arbitrary, the graphical analysis shown below becomse unwieldy at lower thresholds. The reproducer may change this value as required to fit their needs.
Each aggregate data set is plotted in a panel in one of two plots (two panels each for population and economic impacts.)
Data Processing
We load the data into R directly from the bzip file, caching the result to save time.
data = read.csv(bzfile("repdata-data-StormData.csv.bz2"))Check the data
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 6Aggregating event types by poulation health dimensions
evimpact.fatalities <- aggregate(data$FATALITIES ~ data$EVTYPE, FUN=sum)
evimpact.injuries <- aggregate(data$INJURIES ~ data$EVTYPE, FUN=sum)Aggregating event types by economic impacts
evimpact.prop <- aggregate(data$PROPDMG ~ data$EVTYPE, FUN=sum)
evimpact.crop <- aggregate(data$CROPDMG ~ data$EVTYPE, FUN=sum)Stripping out 0 values and finding the top 5%, then choosing only values above this “quantile”, and then sorting in reverse order
evimpact.fatalities <- evimpact.fatalities[!evimpact.fatalities[,2] == 0, ]
evimpact.fatalities.quantile <- quantile(evimpact.fatalities[,2], 0.95)
evimpact.fatalities <- evimpact.fatalities[evimpact.fatalities[,2] > evimpact.fatalities.quantile, ]
evimpact.fatalities <- evimpact.fatalities[ order(-evimpact.fatalities[,2], evimpact.fatalities[,1]), ]
evimpact.injuries <- evimpact.injuries[!evimpact.injuries[,2] == 0, ]
evimpact.injuries.quantile <- quantile(evimpact.injuries[,2], 0.95)
evimpact.injuries <- evimpact.injuries[evimpact.injuries[,2] > evimpact.injuries.quantile, ]
evimpact.injuries <- evimpact.injuries[ order(-evimpact.injuries[,2], evimpact.injuries[,1]), ]
evimpact.crop <- evimpact.crop[!evimpact.crop[,2] == 0, ]
evimpact.crop.quantile <- quantile(evimpact.crop[,2], 0.95)
evimpact.crop <- evimpact.crop[evimpact.crop[,2] > evimpact.crop.quantile, ]
evimpact.crop <- evimpact.crop[ order(-evimpact.crop[,2], evimpact.crop[,1]), ]
evimpact.prop <- evimpact.prop[!evimpact.prop[,2] == 0, ]
evimpact.prop.quantile <- quantile(evimpact.prop[,2], 0.95)
evimpact.prop <- evimpact.prop[evimpact.prop[,2] > evimpact.prop.quantile, ]
evimpact.prop <- evimpact.prop[ order(-evimpact.prop[,2], evimpact.prop[,1]), ]Results
First we plot the population health impacts.
par(mfrow = c(2, 1))
barplot(names.arg=evimpact.fatalities[,1], height=evimpact.fatalities[,2], cex.names=1, las=2, main="Fatalities by Event Type")
barplot(names.arg=evimpact.injuries[,1], height=evimpact.injuries[,2], cex.names=1, las=2, main="Injuries by Event Type")
Next we plot the economic impact.
par(mfrow = c(2, 1), mar=c(10, 6, 1, 1))
barplot(names.arg=evimpact.prop[,1], height=evimpact.prop[,2], cex.names=1, las=2, main="Property Damage ($) by Event Type")
barplot(names.arg=evimpact.crop[,1], height=evimpact.crop[,2], cex.names=1, las=2, main="Crop Damage ($) by Event Type")
It seems that Tornados are, by far, the most impactful events when it comes to human health, and contribute significantly to property damage as well.