Analysis of effects of storms and other weather phenomena in the US
Synopsis
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, property and crop damage. Preventing such outcomes to the extent possible is a key concern.
This data analysis provides information to government or municipal managers who might be responsible for preparing for severe weather events and will need to prioritize resources for different types of events.
The follwing questions are addresed regarding the United States:
- Which types of events are most harmful with respect to population health?
- Which types of events have the greatest economic consequences?
According to the data of the NOAA storm database Tornado’s have by far the most adverse health effects between 1950 and 2011 in the United States. Tornado’s also score highest regarding estimated costs to repair property damage. Drought though seem to cause the hightes estimated costs regarding crop damage.
Data Processing
Description of the source data
To answer the questions an analysis was done on the U.S. National Oceanic and Atmospheric Administration’s (NOAA) storm database. This 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. The events in the database start in the year 1950 and end in November 2011.
The source data is a comma-separated-value file compressed via the bzip2 algorithm to reduce its size. Documentation of the database and variables are available via the National Weather SErvice and the National Climatic Data Center:
Download source and load in R
fileURL <- "https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2"
destfile <- "StormData.csv.bz2"
download.file(fileURL, destfile, method = "curl")
downloaddate <- date()
stormdataRAW <- read.table(destfile, header=T, sep=",")The source data was downloaded as a zipped file on Tue Aug 13 19:04:42 2019 from https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2.
dim(stormdataRAW)## [1] 902297 37The source data contains about a million observations to 37 variables.
Preparation for analysis
The following steps were undertaken to make the data ready for analysis. Details including code can be viewed at the appendix at the end of the report.
Reduce raw source to relevant columns
- Seen the size of the sourcefile the relevant colums were selected for further analysis.
Clean eventtype
add new field: eventtype and put all EVTYPES to uppercase
cleaning typo’s, inconsistent use of singular/plural
Create a dataframe to base analysis of health effects
- reduce to columns eventtype, fatalities and injuries
Convert multiplier and calculate estimated costs
According to the documentation of the data source alphabetical characters used to signify magnitude include “K” for thousands, “M” for millions, and “B” for billions,i.e. 1.55B for $1,550,000,000. “k” and “m” were interpreted as “K” and “M”. For “h” and “H” 100 was taken as a multiplier.
An analysis how to interprete the data which does not comply to characters K, M B was found at https://rstudio-pubs-static.s3.amazonaws.com/58957_37b6723ee52b455990e149edde45e5b6.html
In order to analyse economic effects four new variables were created:
add new variable propmultiplier and transform values from PROPDMGEXP
add new variable cropmultiplier and transform values from CROPDMGEXP
- add new variable propcost = propmultiplier * PROPDMG
add new variable cropcost = cropmultiplier * CROPDMG
Create a dataframe to base analysis of economic effects
- reduce to columns eventtype, propcost and cropcost
Analysis
Weather events most harmful for health
Below we list the top ten of three variables
- sorted descending by total: describes the combined effect (number of fatalities AND injuries)
- sorted descending by fatallities
- sorted descending by injuries
The three lists show that Tornado’s, Thunderstorms/Lightning, (Excesive) Heat and (Flash) Flood are on top of all three lists.
## eventtype FATALITIES INJURIES total
## 573 TORNADO 5633 91346 96979
## 557 THUNDERSTORM WIND 710 9458 10168
## 103 EXCESSIVE HEAT 1922 6575 8497
## 126 FLOOD 476 6791 7267
## 310 LIGHTNING 817 5232 6049
## 212 HEAT 937 2100 3037
## 124 FLASH FLOOD 1035 1800 2835
## 279 ICE STORM 89 1975 2064
## 258 HIGH WIND 293 1471 1764
## 665 WILDFIRE 90 1606 1696## eventtype FATALITIES INJURIES total
## 573 TORNADO 5633 91346 96979
## 103 EXCESSIVE HEAT 1922 6575 8497
## 124 FLASH FLOOD 1035 1800 2835
## 212 HEAT 937 2100 3037
## 310 LIGHTNING 817 5232 6049
## 557 THUNDERSTORM WIND 710 9458 10168
## 402 RIP CURRENT 577 529 1106
## 126 FLOOD 476 6791 7267
## 258 HIGH WIND 293 1471 1764
## 112 EXTREME COLD 288 255 543## eventtype FATALITIES INJURIES total
## 573 TORNADO 5633 91346 96979
## 557 THUNDERSTORM WIND 710 9458 10168
## 126 FLOOD 476 6791 7267
## 103 EXCESSIVE HEAT 1922 6575 8497
## 310 LIGHTNING 817 5232 6049
## 212 HEAT 937 2100 3037
## 279 ICE STORM 89 1975 2064
## 124 FLASH FLOOD 1035 1800 2835
## 665 WILDFIRE 90 1606 1696
## 258 HIGH WIND 293 1471 1764Weather events most harmful for economy
Below we list the top ten of three variables
- sorted descending by total: describes the combined effect (number of property AND crop costs)
- sorted descending by propcost
- sorted descending by cropcost
The three lists show that overall Tornado’s, (Flash) Flood, Hail and Hurricanes result in the highest cost estimates. However, weather effects seem to have different effects on property vs agricultural (crop) cost estimates: Damage to agricultural (crop) is highest due to drought, whereas drought is not in the top ten for property damage.
## eventtype propcost cropcost total
## 573 TORNADO 51690162897 414954710 52105117607
## 126 FLOOD 23490964860 5670823950 29161788810
## 124 FLASH FLOOD 15916911203 1532197150 17449108353
## 181 HAIL 13950269877 3025954650 16976224527
## 265 HURRICANE 11100180010 4020392800 15120572810
## 73 DROUGHT 1046106000 12487566000 13533672000
## 557 THUNDERSTORM WIND 9761179222 1224398700 10985577922
## 665 WILDFIRE 5876463500 402281630 6278745130
## 258 HIGH WIND 4716353745 686301900 5402655645
## 279 ICE STORM 3944928310 72113500 4017041810## eventtype propcost cropcost total
## 573 TORNADO 51690162897 414954710 52105117607
## 126 FLOOD 23490964860 5670823950 29161788810
## 124 FLASH FLOOD 15916911203 1532197150 17449108353
## 181 HAIL 13950269877 3025954650 16976224527
## 265 HURRICANE 11100180010 4020392800 15120572810
## 557 THUNDERSTORM WIND 9761179222 1224398700 10985577922
## 665 WILDFIRE 5876463500 402281630 6278745130
## 258 HIGH WIND 4716353745 686301900 5402655645
## 279 ICE STORM 3944928310 72113500 4017041810
## 587 TROPICAL STORM 2605890550 678846000 3284736550## eventtype propcost cropcost total
## 73 DROUGHT 1046106000 12487566000 13533672000
## 126 FLOOD 23490964860 5670823950 29161788810
## 265 HURRICANE 11100180010 4020392800 15120572810
## 181 HAIL 13950269877 3025954650 16976224527
## 124 FLASH FLOOD 15916911203 1532197150 17449108353
## 112 EXTREME COLD 132385400 1313023000 1445408400
## 557 THUNDERSTORM WIND 9761179222 1224398700 10985577922
## 157 FROST/FREEZE 10480000 1094186000 1104666000
## 222 HEAVY RAIN 694248090 733399800 1427647890
## 258 HIGH WIND 4716353745 686301900 5402655645Disclaimer
As mentioned in the National Climatic Data Center Storm Data FAQ page the National Weather Service does not guarantee the accuracy or validity of the information. In the earlier years of the database there are generally fewer events recorded, most likely due to a lack of good records. More recent years should be considered more complete.
The categorisation of eventtypes is somewhat arbitrary and does not allow a sound analysis. For example, terminology varies from “Heat”, Excessive heat" to “Extreme heat”, etc. In order to deliver good decision material efforts to catalogue eventtypes consistently and reliably with domain knowledge are adviced.
Due to time restraints there was no analysis of missing values made nor there potential influence on the results.
It might be of interest to conduct an analysis on health and economic effect of weather events restricted to the last 5 or 10 years. The data quality might be better and the validity of the results for future budgeting might increase.
Results
Weather events most harmful to health
According to the data of the NOAA storm database Tornado’s have by far the most adverse health effects between 1950 and 2011 in the United States.
Weather events most harmful to economy
Tornado’s also score highest regarding estimated costs to repair property damage. Drought seem to cause the hightes estimated costs regarding crop damage.
Appendix: Code Details
This section offers all coding details of the analysis.
Preparation for analysis
The following steps were undertaken to make the data ready for analysis.
Reduce raw source to relevant columns
- Seen the size of the sourcefile the relevant colums were selected for further analysis.
stormdata <- stormdataRAW[,c("BGN_DATE","TIME_ZONE", "STATE","EVTYPE","FATALITIES", "INJURIES", "PROPDMG", "PROPDMGEXP", "CROPDMG", "CROPDMGEXP")]
dim(stormdata)## [1] 902297 10str(stormdata)## 'data.frame': 902297 obs. of 10 variables:
## $ BGN_DATE : Factor w/ 16335 levels "1/1/1966 0:00:00",..: 6523 6523 4242 11116 2224 2224 2260 383 3980 3980 ...
## $ TIME_ZONE : Factor w/ 22 levels "ADT","AKS","AST",..: 7 7 7 7 7 7 7 7 7 7 ...
## $ 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 ...
## $ 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 ...summary(stormdata)## BGN_DATE TIME_ZONE STATE
## 5/25/2011 0:00:00: 1202 CST :547493 TX : 83728
## 4/27/2011 0:00:00: 1193 EST :245558 KS : 53440
## 6/9/2011 0:00:00 : 1030 MST : 68390 OK : 46802
## 5/30/2004 0:00:00: 1016 PST : 28302 MO : 35648
## 4/4/2011 0:00:00 : 1009 AST : 6360 IA : 31069
## 4/2/2006 0:00:00 : 981 HST : 2563 NE : 30271
## (Other) :895866 (Other): 3631 (Other):621339
## EVTYPE FATALITIES INJURIES
## HAIL :288661 Min. : 0.0000 Min. : 0.0000
## TSTM WIND :219940 1st Qu.: 0.0000 1st Qu.: 0.0000
## THUNDERSTORM WIND: 82563 Median : 0.0000 Median : 0.0000
## TORNADO : 60652 Mean : 0.0168 Mean : 0.1557
## FLASH FLOOD : 54277 3rd Qu.: 0.0000 3rd Qu.: 0.0000
## FLOOD : 25326 Max. :583.0000 Max. :1700.0000
## (Other) :170878
## PROPDMG PROPDMGEXP CROPDMG CROPDMGEXP
## Min. : 0.00 :465934 Min. : 0.000 :618413
## 1st Qu.: 0.00 K :424665 1st Qu.: 0.000 K :281832
## Median : 0.00 M : 11330 Median : 0.000 M : 1994
## Mean : 12.06 0 : 216 Mean : 1.527 k : 21
## 3rd Qu.: 0.50 B : 40 3rd Qu.: 0.000 0 : 19
## Max. :5000.00 5 : 28 Max. :990.000 B : 9
## (Other): 84 (Other): 9head(stormdata)## BGN_DATE TIME_ZONE STATE EVTYPE FATALITIES INJURIES PROPDMG
## 1 4/18/1950 0:00:00 CST AL TORNADO 0 15 25.0
## 2 4/18/1950 0:00:00 CST AL TORNADO 0 0 2.5
## 3 2/20/1951 0:00:00 CST AL TORNADO 0 2 25.0
## 4 6/8/1951 0:00:00 CST AL TORNADO 0 2 2.5
## 5 11/15/1951 0:00:00 CST AL TORNADO 0 2 2.5
## 6 11/15/1951 0:00:00 CST AL TORNADO 0 6 2.5
## PROPDMGEXP CROPDMG CROPDMGEXP
## 1 K 0
## 2 K 0
## 3 K 0
## 4 K 0
## 5 K 0
## 6 K 0Clean eventtype
- add new field: eventtype and put all EVTYPES to uppercase
stormdata$eventtype <- toupper(stormdata$EVTYPE)- cleaning typo’s, inconsistent use of singular/plural
stormdata[stormdata$eventtype == "AVALANCE",]$eventtype <- "AVALANCHE"
stormdata[substr(stormdata$eventtype,1,10) == "COASTAL FL",]$eventtype <- "COASTAL FLOOD"
stormdata[stormdata$eventtype == "COASTALSTORM",]$eventtype <- "COASTAL STORM"
stormdata[stormdata$eventtype == "COLD AND SNOW",]$eventtype <- "COLD"
stormdata[stormdata$eventtype == "COLD TEMPERATURE",]$eventtype <- "COLD"
stormdata[stormdata$eventtype == "COLD WEATHER",]$eventtype <- "COLD"
stormdata[substr(stormdata$eventtype,1,6) == "COLD/W",]$eventtype <- "COLD/WIND"
stormdata[stormdata$eventtype == "DRY MIRCOBURST WINDS",]$eventtype <- "DRY MICROBURST"
stormdata[stormdata$eventtype == "DUST DEVIL",]$eventtype <- "DUST STORM"
stormdata[stormdata$eventtype == "EXTREME COLD/WIND CHILL",]$eventtype <- "EXTREME COLD"
stormdata[substr(stormdata$eventtype,1,11) == "FLASH FLOOD",]$eventtype <- "FLASH FLOOD"
stormdata[substr(stormdata$eventtype,1,12) == " FLASH FLOOD",]$eventtype <- "FLASH FLOOD"
stormdata[stormdata$eventtype == "FLOOD/FLASH FLOOD",]$eventtype <- "FLASH FLOOD"
stormdata[stormdata$eventtype == "FLOODING",]$eventtype <- "FLOOD"
stormdata[stormdata$eventtype == "GUSTY WINDS",]$eventtype <- "GUSTY WIND"
stormdata[stormdata$eventtype == "HEAT WAVES",]$eventtype <- "HEAT WAVE"
stormdata[substr(stormdata$eventtype,1,10) == "HEAVY SNOW",]$eventtype <- "HEAVY SNOW"
stormdata[substr(stormdata$eventtype,1,10) == "HEAVY SURF",]$eventtype <- "HEAVY SURF"
stormdata[substr(stormdata$eventtype,1,9) == "HIGH WIND",]$eventtype <- "HIGH WIND"
stormdata[substr(stormdata$eventtype,1,9) == "HURRICANE",]$eventtype <- "HURRICANE"
stormdata[stormdata$eventtype == "HYPOTHERMIA/EXPOSURE",]$eventtype <- "HYPOTHERMIA"
stormdata[substr(stormdata$eventtype,1,9) == "LIGHTNING",]$eventtype <- "LIGHTNING"
stormdata[stormdata$eventtype == "RECORD/EXCESSIVE HEAT",]$eventtype <- "EXCESSIVE HEAT"
stormdata[stormdata$eventtype == "RECORD HEAT",]$eventtype <- "EXCESSIVE HEAT"
stormdata[stormdata$eventtype == "RECORD COLD",]$eventtype <- "EXTREME COLD"
stormdata[substr(stormdata$eventtype,1,12) == "RIP CURRENTS",]$eventtype <- "RIP CURRENT"
stormdata[stormdata$eventtype == "RIVER FLOODING",]$eventtype <- "RIVER FLOOD"
stormdata[stormdata$eventtype == "SNOW SQUALLS",]$eventtype <- "SNOW SQUALL"
stormdata[stormdata$eventtype == "STRONG WINDS",]$eventtype <- "STRONG WIND"
stormdata[substr(stormdata$eventtype,1,17) == "THUNDERSTORM WIND",]$eventtype <- "THUNDERSTORM WIND"
stormdata[substr(stormdata$eventtype,1,17) == "THUNDERTORM WINDS",]$eventtype <- "THUNDERSTORM WIND"
stormdata[stormdata$eventtype == "TROPICAL STORM GORDON",]$eventtype <- "TROPICAL STORM"
stormdata[substr(stormdata$eventtype,1,5) == "TSTM ",]$eventtype <- "THUNDERSTORM WIND"
stormdata[substr(stormdata$eventtype,1,6) == " TSTM ",]$eventtype <- "THUNDERSTORM WIND"
stormdata[substr(stormdata$eventtype,1,10) == "WILD FIRES",]$eventtype <- "WILDFIRE"
stormdata[substr(stormdata$eventtype,1,10) == "WILD/FORES",]$eventtype <- "WILDFIRE"
stormdata[substr(stormdata$eventtype,1,14) == "WINTER WEATHER",]$eventtype <- "WINTER WEATHER"
stormdata[stormdata$eventtype == "WINTRY MIX",]$eventtype <- "WINTER WEATHER"Create a dataframe to base analysis of health effects
- reduce to columns eventtype, fatalities and injuries
fatal <- aggregate(FATALITIES ~ eventtype, stormdata, FUN ="sum")
injured <- aggregate(INJURIES ~ eventtype, stormdata, FUN ="sum")
healtheffects <- fatal
healtheffects$INJURIES <- injured$INJURIES
dim(healtheffects)## [1] 685 3# reduce to rows where either fatalities or injuries > 0
healtheffects <- healtheffects[(healtheffects$FATALITIES > 0) | (healtheffects$INJURIES >0), ]
dim(healtheffects)## [1] 135 3Convert multiplier and calculate estimated costs
In order to analyse economic effects four new variables were created:
- add new variable propmultiplier and transform values from PROPDMGEXP
stormdata$propmultiplier <- 10
stormdata$propmultiplier <- as.numeric(stormdata$propmultiplier)
#plot(stormdata$PROPDMGEXP)
table(stormdata$PROPDMGEXP)##
## - ? + 0 1 2 3 4 5
## 465934 1 8 5 216 25 13 4 4 28
## 6 7 8 B h H K m M
## 4 5 1 40 1 6 424665 7 11330stormdata[stormdata$PROPDMGEXP == "",]$propmultiplier <- 0
stormdata[stormdata$PROPDMGEXP == "-",]$propmultiplier <- 0
stormdata[stormdata$PROPDMGEXP == "?",]$propmultiplier <- 0
stormdata[stormdata$PROPDMGEXP == "+",]$propmultiplier <- 1
stormdata[stormdata$PROPDMGEXP == "B",]$propmultiplier <- 10000000
stormdata[stormdata$PROPDMGEXP == "h",]$propmultiplier <- 100
stormdata[stormdata$PROPDMGEXP == "H",]$propmultiplier <- 100
stormdata[stormdata$PROPDMGEXP == "K",]$propmultiplier <- 1000
stormdata[stormdata$PROPDMGEXP == "m",]$propmultiplier <- 1000000
stormdata[stormdata$PROPDMGEXP == "M",]$propmultiplier <- 1000000
options(scipen=999)
table(stormdata$propmultiplier)##
## 0 1 10 100 1000 1000000 10000000
## 465943 5 300 7 424665 11337 40- add new variable cropmultiplier and transform values from CROPDMGEXP
stormdata$cropmultiplier <- 10
stormdata$cropmultiplier <- as.numeric(stormdata$cropmultiplier)
#plot(stormdata$CROPDMGEXP)
table(stormdata$CROPDMGEXP)##
## ? 0 2 B k K m M
## 618413 7 19 1 9 21 281832 1 1994stormdata[stormdata$CROPDMGEXP == "",]$cropmultiplier <- 0
stormdata[stormdata$CROPDMGEXP == "?",]$cropmultiplier <- 0
stormdata[stormdata$CROPDMGEXP == "B",]$cropmultiplier <- 10000000
stormdata[stormdata$CROPDMGEXP == "k",]$cropmultiplier <- 1000
stormdata[stormdata$CROPDMGEXP == "K",]$cropmultiplier <- 1000
stormdata[stormdata$CROPDMGEXP == "m",]$cropmultiplier <- 1000000
stormdata[stormdata$CROPDMGEXP == "M",]$cropmultiplier <- 1000000
options(scipen=999)
table(stormdata$cropmultiplier)##
## 0 10 1000 1000000 10000000
## 618420 20 281853 1995 9- add new variable propcost = propmultiplier * PROPDMG
- add new variable cropcost = cropmultiplier * CROPDMG
#dim(stormdata)
stormdata$propcost <- stormdata$propmultiplier * stormdata$PROPDMG
stormdata$cropcost <- stormdata$cropmultiplier * stormdata$CROPDMG
#dim(stormdata)Create a dataframe to base analysis of economic effects
- reduce to columns eventtype, propcost and cropcost
prop <- aggregate(propcost ~ eventtype, stormdata, FUN ="sum")
crop <- aggregate(cropcost ~ eventtype, stormdata, FUN ="sum")
economiceffects <- prop
economiceffects$cropcost <- crop$cropcost
dim(economiceffects)## [1] 685 3Analysis
Weather events most harmful for health
healtheffects$total <- healtheffects$FATALITIES + healtheffects$INJURIES
head(healtheffects[order(-healtheffects$total),], 10)## eventtype FATALITIES INJURIES total
## 573 TORNADO 5633 91346 96979
## 557 THUNDERSTORM WIND 710 9458 10168
## 103 EXCESSIVE HEAT 1922 6575 8497
## 126 FLOOD 476 6791 7267
## 310 LIGHTNING 817 5232 6049
## 212 HEAT 937 2100 3037
## 124 FLASH FLOOD 1035 1800 2835
## 279 ICE STORM 89 1975 2064
## 258 HIGH WIND 293 1471 1764
## 665 WILDFIRE 90 1606 1696head(healtheffects[order(-healtheffects$FATALITIES),], 10)## eventtype FATALITIES INJURIES total
## 573 TORNADO 5633 91346 96979
## 103 EXCESSIVE HEAT 1922 6575 8497
## 124 FLASH FLOOD 1035 1800 2835
## 212 HEAT 937 2100 3037
## 310 LIGHTNING 817 5232 6049
## 557 THUNDERSTORM WIND 710 9458 10168
## 402 RIP CURRENT 577 529 1106
## 126 FLOOD 476 6791 7267
## 258 HIGH WIND 293 1471 1764
## 112 EXTREME COLD 288 255 543head(healtheffects[order(-healtheffects$INJURIES),], 10)## eventtype FATALITIES INJURIES total
## 573 TORNADO 5633 91346 96979
## 557 THUNDERSTORM WIND 710 9458 10168
## 126 FLOOD 476 6791 7267
## 103 EXCESSIVE HEAT 1922 6575 8497
## 310 LIGHTNING 817 5232 6049
## 212 HEAT 937 2100 3037
## 279 ICE STORM 89 1975 2064
## 124 FLASH FLOOD 1035 1800 2835
## 665 WILDFIRE 90 1606 1696
## 258 HIGH WIND 293 1471 1764#In order to be able to plot the averse health effects in the result section, we need to prepare the data:
#limit the data for plot
plotdataH <- head(healtheffects[order(-healtheffects$total),], 10)
plotdataH <- plotdataH[,1:3]
#make wide format to long format in order to plot the results
library(reshape2)
head(plotdataH)## eventtype FATALITIES INJURIES
## 573 TORNADO 5633 91346
## 557 THUNDERSTORM WIND 710 9458
## 103 EXCESSIVE HEAT 1922 6575
## 126 FLOOD 476 6791
## 310 LIGHTNING 817 5232
## 212 HEAT 937 2100plotdataH <- melt(plotdataH)## Using eventtype as id variableshead(plotdataH)## eventtype variable value
## 1 TORNADO FATALITIES 5633
## 2 THUNDERSTORM WIND FATALITIES 710
## 3 EXCESSIVE HEAT FATALITIES 1922
## 4 FLOOD FATALITIES 476
## 5 LIGHTNING FATALITIES 817
## 6 HEAT FATALITIES 937Weather events most harmful for economy
economiceffects$total <- economiceffects$propcost + economiceffects$cropcost
head(economiceffects[order(-economiceffects$total),], 10)## eventtype propcost cropcost total
## 573 TORNADO 51690162897 414954710 52105117607
## 126 FLOOD 23490964860 5670823950 29161788810
## 124 FLASH FLOOD 15916911203 1532197150 17449108353
## 181 HAIL 13950269877 3025954650 16976224527
## 265 HURRICANE 11100180010 4020392800 15120572810
## 73 DROUGHT 1046106000 12487566000 13533672000
## 557 THUNDERSTORM WIND 9761179222 1224398700 10985577922
## 665 WILDFIRE 5876463500 402281630 6278745130
## 258 HIGH WIND 4716353745 686301900 5402655645
## 279 ICE STORM 3944928310 72113500 4017041810head(economiceffects[order(-economiceffects$propcost),], 10)## eventtype propcost cropcost total
## 573 TORNADO 51690162897 414954710 52105117607
## 126 FLOOD 23490964860 5670823950 29161788810
## 124 FLASH FLOOD 15916911203 1532197150 17449108353
## 181 HAIL 13950269877 3025954650 16976224527
## 265 HURRICANE 11100180010 4020392800 15120572810
## 557 THUNDERSTORM WIND 9761179222 1224398700 10985577922
## 665 WILDFIRE 5876463500 402281630 6278745130
## 258 HIGH WIND 4716353745 686301900 5402655645
## 279 ICE STORM 3944928310 72113500 4017041810
## 587 TROPICAL STORM 2605890550 678846000 3284736550head(economiceffects[order(-economiceffects$cropcost),], 10)## eventtype propcost cropcost total
## 73 DROUGHT 1046106000 12487566000 13533672000
## 126 FLOOD 23490964860 5670823950 29161788810
## 265 HURRICANE 11100180010 4020392800 15120572810
## 181 HAIL 13950269877 3025954650 16976224527
## 124 FLASH FLOOD 15916911203 1532197150 17449108353
## 112 EXTREME COLD 132385400 1313023000 1445408400
## 557 THUNDERSTORM WIND 9761179222 1224398700 10985577922
## 157 FROST/FREEZE 10480000 1094186000 1104666000
## 222 HEAVY RAIN 694248090 733399800 1427647890
## 258 HIGH WIND 4716353745 686301900 5402655645# In order to be able to plot the averse economic effects in the result section, we need to prepare the data:
# limit the data for plot
plotdataE <- head(economiceffects[order(-economiceffects$total),], 10)
plotdataE <- plotdataE[,1:3]
#make wide format to long format in order to plot the results
#library(reshape2)
#head(plotdataE)
plotdataE <- melt(plotdataE)## Using eventtype as id variables#head(plotdataE)Results
Weather events most harmful to health
library(ggplot2)
p1 <- ggplot(plotdataH, aes(x=reorder(eventtype,-value), y=value, fill=variable))
p1 <- p1 + geom_bar(stat='identity')
p1 <- p1 + theme_bw() + theme(axis.text.x=element_text(angle=45, hjust=1))
p1 <- p1 + labs(y="Number of fatalities/injuries", x = "Eventtype", fill="")
p1 <- p1 + ggtitle("Weather events most harmful to health (Total Top 10)")
p1 + scale_fill_manual(values = c("Black", "Red"))Weather events most harmful to economy
p2 <- ggplot(plotdataE, aes(x=reorder(eventtype,-value), y=value, fill=variable))
p2 <- p2 + geom_bar(stat='identity')
p2 <- p2 + theme_bw() + theme(axis.text.x=element_text(angle=45, hjust=1))
p2 <- p2 + labs(y="Sum of estimated costs", x = "Eventtype", fill="")
p2 <- p2 + ggtitle("Weather events most harmful to economy (Total Top 10)")
p2 + scale_fill_manual(values = c("Steelblue", "Darkgreen"))Software Environment:
sessionInfo()## R version 3.6.1 (2019-07-05)
## Platform: x86_64-w64-mingw32/x64 (64-bit)
## Running under: Windows 10 x64 (build 18362)
##
## Matrix products: default
##
## locale:
## [1] LC_COLLATE=German_Austria.1252 LC_CTYPE=German_Austria.1252
## [3] LC_MONETARY=German_Austria.1252 LC_NUMERIC=C
## [5] LC_TIME=German_Austria.1252
##
## attached base packages:
## [1] stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] ggplot2_3.2.0 reshape2_1.4.3
##
## loaded via a namespace (and not attached):
## [1] Rcpp_1.0.1 knitr_1.23 magrittr_1.5 tidyselect_0.2.5
## [5] munsell_0.5.0 colorspace_1.4-1 R6_2.4.0 rlang_0.4.0
## [9] stringr_1.4.0 plyr_1.8.4 dplyr_0.8.3 tools_3.6.1
## [13] grid_3.6.1 gtable_0.3.0 xfun_0.8 withr_2.1.2
## [17] htmltools_0.3.6 assertthat_0.2.1 lazyeval_0.2.2 digest_0.6.20
## [21] tibble_2.1.3 crayon_1.3.4 purrr_0.3.2 glue_1.3.1
## [25] evaluate_0.14 rmarkdown_1.14 labeling_0.3 stringi_1.4.3
## [29] compiler_3.6.1 pillar_1.4.2 scales_1.0.0 pkgconfig_2.0.2This analysis was a project for the course “Reproducible Research” from Johns Hopkins on Coursera.