NOAA Storm Database Analysis (2nd Assignment)
Ferran Martí
Thursday, May 14, 2015
THE EFFECT OF CLIMATIC EVENTS IN THE US: HEALTH AND ECONOMIC CONSEQUENCES
SYNOPSIS
Severe climatic events such as tornados or typhoons are one of the biggest concerns for the people living in the US, as they can inflict huge economic and personal damage. In this report, the aim is to analyse which of those climatic events are more harmful, regarding human health and regarding economic damage. To do it, we use data from the US National Weather Service. In our findings, we observe that tornados are the worst events in terms of human health, causing the higher number of casualties and injuries, whereas floods are the worst events in terms of economic damage.
LOADING THE DATA
Before starting, it is necessary to load the packages dplyr, knitr and ggplot2 on R:
library(dplyr)##
## Attaching package: 'dplyr'
##
## The following object is masked from 'package:stats':
##
## filter
##
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(knitr)
library(ggplot2)Data (https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2) IS avaible on-line as well as details on this data (https://d396qusza40orc.cloudfront.net/repdata%2Fpeer2_doc%2Fpd01016005curr.pdf).
You can either download the file directly to R, or download on the PC and then load it to R. Here, I apply the second option. So first, use the link I have provided before, and then put the file in the same directory as the working directory in R.
data <- read.csv("C:/Users/ftorrent/Desktop/Data Science Track1/Coursera/Reproducible Research/Assignment 2/repdata-data-StormData.csv/repdata-data-StormData.csv")
dim(data)## [1] 902297 37As we can observe, there are 902297 observations and 37 variables.
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 ...Here we have a list of the variables in our dataset.
PROCESSING THE DATA
As we can recall, there are two different questions to answer in this assignment:
Across the United States, which types of events (as indicated in the EVTYPE variable) are most harmful with respect to population health?
Across the United States, which types of events have the greatest economic consequences?
So, according to those questions, we can clean the dataset we have, so it gets smaller and faster to compute. Therefore, for this analysis we will use (and keep) the variables:
- State (STATE)
- Date (BGN_DATE)
- Type of event (EVTYPE)
- Number of fatalities (FATALITIES)
- Number of injuries (INJURIES)
- Property damage (PROPDMG)
- Property damage units, either K or M (PROPDMGEXP)
- Crop damage (CROPDMG)
- Crop damage units, either K or M (CROPDMGEXP)
The fatalities and injuries variables inform us about the health damage caused by the event, whereas the last four variables inform about the material damage caused.
data<-select(data, STATE, BGN_DATE, EVTYPE, FATALITIES, INJURIES, PROPDMG, PROPDMGEXP, CROPDMG, CROPDMGEXP)Now, let’s format the variables on the dataset. First, we set the date variable on its proper category:
data$date = as.Date(data$BGN_DATE, format = "%m/%d/%Y")Then we also want to merge the PROPDMG and PROPDMGEXP variable and the CROPDMG and CROPDMGEXP, so we get only one complete number for each observation. By using the levels formula, we can see how many different “measures” are there in PROPDMGEXP and CROPDMGEXP:
levels(data$CROPDMGEXP)## [1] "" "?" "0" "2" "B" "k" "K" "m" "M"levels(data$PROPDMGEXP)## [1] "" "-" "?" "+" "0" "1" "2" "3" "4" "5" "6" "7" "8" "B" "h" "H" "K"
## [18] "m" "M"Therefore, we have to merge this “units” with the numbers provided in PROPDMG and CROPDMG:
data$PROPDMGEXP<- as.character(data$PROPDMGEXP)
data$propdamagemultiplier<-1
data[data$PROPDMGEXP == "H", ]$propdamagemultiplier <- 100
data[data$PROPDMGEXP == "K", ]$propdamagemultiplier <- 1000
data[data$PROPDMGEXP == "M", ]$propdamagemultiplier <- 1000000
data[data$PROPDMGEXP == "B", ]$propdamagemultiplier <- 1000000000
data[data$PROPDMGEXP == "h", ]$propdamagemultiplier <- 100
data[data$PROPDMGEXP == "m", ]$propdamagemultiplier <- 1000000
data$totalpropertydamage <- data$PROPDMG*data$propdamagemultiplier
data$CROPDMGEXP<- as.character(data$CROPDMGEXP)
data$cropdamagemultiplier<-1
data[data$CROPDMGEXP == "K", ]$cropdamagemultiplier <- 1000
data[data$CROPDMGEXP == "M", ]$cropdamagemultiplier <- 1000000
data[data$CROPDMGEXP == "B", ]$cropdamagemultiplier <- 1000000000
data[data$CROPDMGEXP == "k", ]$cropdamagemultiplier <- 1000
data[data$CROPDMGEXP == "m", ]$cropdamagemultiplier <- 1000000
data$totalcropdamage <- data$CROPDMG*data$cropdamagemultiplierI clean again the dataset, dropping the variables I won’t use for my analysis:
data<-select(data, STATE, EVTYPE, FATALITIES, INJURIES, date, totalpropertydamage, totalcropdamage)Now, to have a look closely to the dataset and see its completeness, we can observe which quantity of observations are there every year:
data$year <- as.integer(strftime(data$date,"%Y"))
hist(data$year, main = "N° of events per year", xlab="Year")
As we can observe,the events in the database start in the year 1950 and end in mid 2011. In the earlier years of the database there are fewer events recorded, most likely due to a lack of good records or processing information techniques. More recent years should be considered more complete. Another interpretation could be that the number of events per year have been increasing due to global warming effects or other long-term climatic issues. In 2011 the number of events drops, but this is because the dataset is incomplete on that year, therefore we should focus on the increasing pattern until 2010.
RESULTS
HARMFUL EVENTS FOR POPULATION HEALTH
There are two variables indicating population health: injuries and fatalities. As there are 985 different types of events, we select the top 15 for injuries and for fatalities:
FatilitiesByEvent<-aggregate(FATALITIES ~ EVTYPE,data=data,sum)
FatilitiesByEventSorted <- FatilitiesByEvent[order(-FatilitiesByEvent$FATALITIES),]
FatilitiesByEventTOP15<-FatilitiesByEventSorted[1:15, ]
FatilitiesByEventTOP15## EVTYPE FATALITIES
## 834 TORNADO 5633
## 130 EXCESSIVE HEAT 1903
## 153 FLASH FLOOD 978
## 275 HEAT 937
## 464 LIGHTNING 816
## 856 TSTM WIND 504
## 170 FLOOD 470
## 585 RIP CURRENT 368
## 359 HIGH WIND 248
## 19 AVALANCHE 224
## 972 WINTER STORM 206
## 586 RIP CURRENTS 204
## 278 HEAT WAVE 172
## 140 EXTREME COLD 160
## 760 THUNDERSTORM WIND 133InjuriesByEvent<-aggregate(INJURIES ~ EVTYPE,data=data,sum)
InjuriesByEventSorted <- InjuriesByEvent[order(-InjuriesByEvent$INJURIES),]
InjuriesByEventTOP15<-InjuriesByEventSorted[1:15, ]
InjuriesByEventTOP15## EVTYPE INJURIES
## 834 TORNADO 91346
## 856 TSTM WIND 6957
## 170 FLOOD 6789
## 130 EXCESSIVE HEAT 6525
## 464 LIGHTNING 5230
## 275 HEAT 2100
## 427 ICE STORM 1975
## 153 FLASH FLOOD 1777
## 760 THUNDERSTORM WIND 1488
## 244 HAIL 1361
## 972 WINTER STORM 1321
## 411 HURRICANE/TYPHOON 1275
## 359 HIGH WIND 1137
## 310 HEAVY SNOW 1021
## 957 WILDFIRE 911If we add those two, we can plot a graph indicating the type of event that has caused more damage, either in fatalities or in injuries:
data$populationhealth <- (data$FATALITIES + data$INJURIES)
PopulationhealthByEvent<-aggregate(populationhealth ~ EVTYPE,data=data,sum)
PopulationhealthByEventSorted <- PopulationhealthByEvent[order(-PopulationhealthByEvent$populationhealth),]
PopulationhealthByEventTOP15<-PopulationhealthByEventSorted[1:15, ]
PopulationhealthByEventTOP15## EVTYPE populationhealth
## 834 TORNADO 96979
## 130 EXCESSIVE HEAT 8428
## 856 TSTM WIND 7461
## 170 FLOOD 7259
## 464 LIGHTNING 6046
## 275 HEAT 3037
## 153 FLASH FLOOD 2755
## 427 ICE STORM 2064
## 760 THUNDERSTORM WIND 1621
## 972 WINTER STORM 1527
## 359 HIGH WIND 1385
## 244 HAIL 1376
## 411 HURRICANE/TYPHOON 1339
## 310 HEAVY SNOW 1148
## 957 WILDFIRE 986barplot (height=PopulationhealthByEventTOP15$populationhealth, names.arg=PopulationhealthByEventTOP15$EVTYPE[1:15], las=2, cex.names=0.6,
col = rainbow (15, start=0, end=0.5))
title (main = "N° of injuries and fatalities across events", line=1)
title (ylab = "Total number of Injuries and Fatalities", line=4, cex.lab=0.5)
As we can see, tornados are, by far, the worst event in terms of human health, as it accounts for more than 5500 fatalities and 90000 injuries. Following tornados, massive heart and TSTM wind are the most harmful events for human health.
ECONOMIC DAMAGES
To see the economic damages, we can add the total crop damage and the total property damage. Recall we have modified those variables before, so they become totally numeric and we can operate with them. Let’s see first the top 15 events that cause more property damage, then the top 15 that cause crop damage and finally the top 15 adding those two variables:
PropertydamageByEvent<-aggregate(totalpropertydamage ~ EVTYPE,data=data,sum)
PropertydamageByEventByEventSorted <- PropertydamageByEvent[order(-PropertydamageByEvent$totalpropertydamage),]
PropertydamageByEventByEventTOP15<-PropertydamageByEventByEventSorted[1:15, ]
PropertydamageByEventByEventTOP15## EVTYPE totalpropertydamage
## 170 FLOOD 144657709807
## 411 HURRICANE/TYPHOON 69305840000
## 834 TORNADO 56937160779
## 670 STORM SURGE 43323536000
## 153 FLASH FLOOD 16140812067
## 244 HAIL 15732267543
## 402 HURRICANE 11868319010
## 848 TROPICAL STORM 7703890550
## 972 WINTER STORM 6688497251
## 359 HIGH WIND 5270046295
## 590 RIVER FLOOD 5118945500
## 957 WILDFIRE 4765114000
## 671 STORM SURGE/TIDE 4641188000
## 856 TSTM WIND 4484928495
## 427 ICE STORM 3944927860CropdamageByEvent<-aggregate(totalcropdamage ~ EVTYPE,data=data,sum)
CropdamageByEventByEventSorted <- CropdamageByEvent[order(-CropdamageByEvent$totalcropdamage),]
CropdamageByEventByEventTOP15<-CropdamageByEventByEventSorted[1:15, ]
CropdamageByEventByEventTOP15## EVTYPE totalcropdamage
## 95 DROUGHT 13972566000
## 170 FLOOD 5661968450
## 590 RIVER FLOOD 5029459000
## 427 ICE STORM 5022113500
## 244 HAIL 3025954473
## 402 HURRICANE 2741910000
## 411 HURRICANE/TYPHOON 2607872800
## 153 FLASH FLOOD 1421317100
## 140 EXTREME COLD 1292973000
## 212 FROST/FREEZE 1094086000
## 290 HEAVY RAIN 733399800
## 848 TROPICAL STORM 678346000
## 359 HIGH WIND 638571300
## 856 TSTM WIND 554007350
## 130 EXCESSIVE HEAT 492402000Here it is interesting to see that draughts are more than twice as harmful as floods for the crops. However, draughts don’t cause any property damage, therefore when we add property and crop damage this variable falls down the “top” harmful events.
data$totaleconomicimpact<-data$totalcropdamage+data$totalpropertydamage
TotaldamageByEvent<-aggregate(totaleconomicimpact ~ EVTYPE,data=data,sum)
TotaldamageByEventByEventSorted <- TotaldamageByEvent[order(-TotaldamageByEvent$totaleconomicimpact),]
TotaldamageByEventByEventTOP15<-TotaldamageByEventByEventSorted[1:15, ]
TotaldamageByEventByEventTOP15## EVTYPE totaleconomicimpact
## 170 FLOOD 150319678257
## 411 HURRICANE/TYPHOON 71913712800
## 834 TORNADO 57352114049
## 670 STORM SURGE 43323541000
## 244 HAIL 18758222016
## 153 FLASH FLOOD 17562129167
## 95 DROUGHT 15018672000
## 402 HURRICANE 14610229010
## 590 RIVER FLOOD 10148404500
## 427 ICE STORM 8967041360
## 848 TROPICAL STORM 8382236550
## 972 WINTER STORM 6715441251
## 359 HIGH WIND 5908617595
## 957 WILDFIRE 5060586800
## 856 TSTM WIND 5038935845And we plot the graph, but before let’s transform the total amounts to billions, so it is easier to read:
TotaldamageByEventByEventTOP15$billions<-(TotaldamageByEventByEventTOP15$totaleconomicimpact)/1000000000
barplot (height=TotaldamageByEventByEventTOP15$billions, names.arg=TotaldamageByEventByEventTOP15$EVTYPE[1:15], las=2, cex.names=0.6,
col = rainbow (15, start=0, end=0.5))
title (main = "Economic damage", line=1)
title (ylab = "Total Damage (Billions USD)", line=3, cex.lab=0.8)
So we can observe that, in economic terms, a flood is the worst thing that can happen to the US, doing twice as damage that a typhoon and almost three times the damage caused by a tornado.
CONCLUSION
In terms of human health, a tornado is the worst event, being the most harmful event that can happen in the US. When evaluating the damage caused by a climatic event in terms of economic damage, floods have caused a damage of more than 150 billion USD since 1950. Typhoons account for 70 billion damage and tornados for 60 billion economic damage in the US economy.