Health and Economic Impact of Weather Events in the US
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, and property damage, and preventing such outcomes to the extent possible is a key concern.
This project involves exploring 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.
Synopsis
The analysis on the storm event database revealed that tornadoes are the most dangerous weather event to the population health. The second most dangerous event type is the excessive heat. The economic impact of weather events was also analyzed. Flash floods and thunderstorm winds caused billions of dollars in property damages between 1950 and 2011. The largest crop damage caused by drought, followed by flood and hails.
Data Processing
The analysis was performed on Storm Events Database, provided by National Climatic Data Center. The data is from a comma-separated-value file available here. There is also some documentation of the data available here.
The first step is to read the data into a data frame.
dat <- read.csv('data/repdata_data_StormData.csv.bz2')Before the analysis, the data need some preprocessing. Event types
don’t have a specific format. For instance, there are events with types
Frost/Freeze, FROST/FREEZE and
FROST\\FREEZE which obviously refer to the same type of
event.
# number of unique event types
length(unique(dat$EVTYPE))## [1] 985event_types <- dat$EVTYPE
event_types <- gsub("[[:blank:][:punct:]+]", " ", event_types)
#Update the data frame
dat$EVTYPE <- event_typesNo further data preprocessing was performed although the event type
field can be processed further to merge event types such as
tstm wind and thunderstorm wind. After the
cleaning, as expected, the number of unique event types reduce
significantly. For further analysis, the cleaned event types are
used.
Dangerous Events with respect to Population Health
To find the event types that are most harmful to population health, the number of casualties are aggregated by the event type.
library(dplyr)##
## Adjuntando el paquete: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unioneventype_fatalities <- as.data.frame(dat %>% group_by(EVTYPE) %>% summarise(fatalities = sum(FATALITIES), injuries = sum(INJURIES)))#We yield the top 10 most fatal event types
head(eventype_fatalities[order(eventype_fatalities$fatalities, decreasing = TRUE), c('EVTYPE', "fatalities")], 10)## EVTYPE fatalities
## 810 TORNADO 5633
## 124 EXCESSIVE HEAT 1903
## 151 FLASH FLOOD 978
## 268 HEAT 937
## 446 LIGHTNING 816
## 830 TSTM WIND 504
## 167 FLOOD 470
## 564 RIP CURRENT 368
## 337 HIGH WIND 248
## 19 AVALANCHE 224#We yield the top 10 event types which caused more injuries
head(eventype_fatalities[order(eventype_fatalities$injuries, decreasing = TRUE), c("EVTYPE","injuries")], 10)## EVTYPE injuries
## 810 TORNADO 91346
## 830 TSTM WIND 6957
## 167 FLOOD 6789
## 124 EXCESSIVE HEAT 6525
## 446 LIGHTNING 5230
## 268 HEAT 2100
## 417 ICE STORM 1975
## 151 FLASH FLOOD 1777
## 740 THUNDERSTORM WIND 1488
## 238 HAIL 1361Economic Effects of Weather Events
To analyze the impact of weather events on the economy, available property damage and crop damage reportings/estimates were used.
In the raw data, the property damage is represented with two fields,
a number PROPDMG in dollars and the exponent
PROPDMGEXP. Similarly, the crop damage is represented using
two fields, CROPDMG and CROPDMGEXP. The first
step in the analysis is to calculate the property and crop damage for
each event.
#Let's first create a function which will return the exponent digit given a letter indicating the exponent
exp_function <- function(x){
if(x %in% c('k', 'K')){
return(3)
}
else if(x %in% c('M', 'm')){
return(6)
}
else if(x %in% c('B', 'b')){
return(9)
}
else if(x %in% c('h','H')){
return(2)
}
else if(!is.na(as.numeric(x))){
return(as.numeric(x))
}
else if(x %in% c('', '-', '?', '+')){
return(0)
}
else{
stop('Invalid entry')
}
}
#Now let's add this changes to our original data file
prop_dmg_exp <- sapply(dat$PROPDMGEXP, FUN = exp_function)## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coercióndat$prop_dmg <- dat$PROPDMG * (10**prop_dmg_exp)
crop_dmg_exp <- sapply(dat$CROPDMGEXP, FUN = exp_function)## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coerción
## Warning in FUN(X[[i]], ...): NAs introducidos por coercióndat$crop_dmg <- dat$CROPDMG * (10**crop_dmg_exp)
#Next, let's calculate the total economic loss per event type
econ_loss <- as.data.frame(dat %>% group_by(EVTYPE) %>% summarise(crop_loss = sum(crop_dmg), prop_loss = sum(prop_dmg)))#We yield the top 10 event types that caused the highest crop losses
head(econ_loss[order(econ_loss$crop_loss, decreasing = TRUE),c('EVTYPE','crop_loss')], 10)## EVTYPE crop_loss
## 91 DROUGHT 13972566000
## 167 FLOOD 5661968450
## 568 RIVER FLOOD 5029459000
## 417 ICE STORM 5022113500
## 238 HAIL 3025954473
## 379 HURRICANE 2741910000
## 387 HURRICANE TYPHOON 2607872800
## 151 FLASH FLOOD 1421317100
## 132 EXTREME COLD 1292973000
## 196 FROST FREEZE 1094086000#We now yield the top 10 event types which caused the highest property losses
head(econ_loss[order(econ_loss$prop_loss, decreasing = TRUE),c('EVTYPE','prop_loss')], 10)## EVTYPE prop_loss
## 167 FLOOD 144657709807
## 387 HURRICANE TYPHOON 69305840000
## 810 TORNADO 56947380677
## 643 STORM SURGE 43323536000
## 151 FLASH FLOOD 16822673979
## 238 HAIL 15735267513
## 379 HURRICANE 11868319010
## 823 TROPICAL STORM 7703890550
## 941 WINTER STORM 6688497251
## 337 HIGH WIND 5270046295Results
Health impact of weather events
The following plot shows top dangerous weather event types. Here, we will only consider the op 10 most fatal disasters type
library(ggplot2)
library(gridExtra)##
## Adjuntando el paquete: 'gridExtra'## The following object is masked from 'package:dplyr':
##
## combineeventype_fatalities_filter <- eventype_fatalities[order(eventype_fatalities$fatalities, decreasing = TRUE),]
eventype_injuries_filter <- eventype_fatalities[order(eventype_fatalities$injuries, decreasing = TRUE),]
p1 <- ggplot(data = head(eventype_fatalities_filter,10), aes(x = EVTYPE, y = fatalities)) + geom_bar(stat = 'identity') + labs(x = 'Event type', y = 'Number of fatalities', title = 'Fatalities as a function of the Event type') + coord_flip()
p2 <- ggplot(data = head(eventype_injuries_filter,10), aes(x = EVTYPE, y = injuries)) + geom_bar(stat = 'identity') + labs(x = 'Event type', y = 'Number of injuries', title = 'Injuries as a function of the Event type') + coord_flip()
grid.arrange(p1, p2, top="Top deadly weather events in the US (1950-2011)")
Economic impact of weather events
The following plot shows the most severe weather event types with respect to economic cost that they have costed since 1950s.
econ_loss_prop_filter <- head(econ_loss[order(econ_loss$prop_loss, decreasing = TRUE),c('EVTYPE','prop_loss')], 10)
econ_loss_crop_filter <- head(econ_loss[order(econ_loss$crop_loss, decreasing = TRUE),c('EVTYPE','crop_loss')], 10)
p1 <- ggplot(data = econ_loss_prop_filter, aes(x = EVTYPE, y = prop_loss)) + geom_bar(stat = 'identity') + labs(x = 'Event type', y = 'Porperty losses', title = 'Property losses as a function of the Event type') + coord_flip()
p2 <- ggplot(data = head(econ_loss_crop_filter,10), aes(x = EVTYPE, y = crop_loss)) + geom_bar(stat = 'identity') + labs(x = 'Event type', y = 'Crop losses', title = 'Crop losses as a function of the Event type') + coord_flip()
grid.arrange(p1, p2, top="Weather costs to the US economy (1950-2011)")
Property damages are given in logarithmic scale due to large range of
values. The data shows that flash floods and thunderstorm winds cost the
largest property damages among weather-related natural diseasters. Note
that, due to untidy nature of the available data, type
flood and flash flood are separate values and
should be merged for more accurate data-driven conclusions.
The most severe weather event in terms of crop damage is the drought. In the last half century, the drought has caused more than 10 billion dollars damage. Other severe crop-damage-causing event types are floods and hails.