Health and Economic Impact of Storms
Charles Chisanga
25th December 2015
### About
This is the second peer assessment project for the Reproducible Research course under Data Science specialization. The purpose of the project was to determine which storm event or events had the most significant economic and health effects on the US. Storm Data is an official publication of the National Oceanic and Atmospheric Administration (NOAA) which documents the occurrence of storms and other significant weather phenomena having sufficient intensity to cause loss of life, injuries, significant property damage, and/or disruption to commerce.
# Synopsis
Severe weather has serious economic and health impacts, causing property damage, crop damage, injury and even death. The purpose of this assignment was to determine which weather types had the greatest economic and health effects. Economic effects were operationalized as the degree of property and crop damage. Health effects were operationalized as number of fatalities and injuries.
Data was taken from the U.S. National Oceanic and Atmospheric Administration’s (NOAA) storm database and analyzed using R, Rstudio and knitr package. The raw data was subjected to preprocessing prior to the analysis.
The report begins with initial data processing followed by a subsequent analysis with the most important results plotted. Property damages are given in logarithmic scale due to large range of values. The report ends with results and briefly discuss. The economic impact of weather events was also analyzed.
## Data Processing
Event data prior to 1996 was incomplete; it only contained Tornado, Thunderstorm, Wind and Hail event types, while data in 1996 and after contains all 48 event types that are in current use. The analysis was performed on Storm Data downloaded from the National Climatic Data Center (url: https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2). There is also some documentation of the data available at url: https://d396qusza40orc.cloudfront.net/repdata%2Fpeer2_doc%2Fpd01016005curr.pdf.
The first step was to read the data into a data frame.
storm <- read.csv(("/home/cchisanga/DataScienceSpCourseCoursera/5.Reproducible_Research/PeerAssignment2/repdata-data-StormData.csv.bz2"), strip.white = TRUE)
# head(storm)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(storm$EVTYPE))## [1] 985# translate all letters to lowercase
event_types <- tolower(storm$EVTYPE)
# replace all punct. characters with a space
event_types <- gsub("[[:blank:][:punct:]+]", " ", event_types)
length(unique(event_types))## [1] 874# update the data frame
storm$EVTYPE <- event_typesNo further data preprocessing was performed although the event type field can be processed further to merge event types such as tstm wind and thunderstorm wind. After the cleaning, as expected, the number of unique event types reduce significantly. For further analysis, the cleaned event types are used.
# Dangerous Events with respect to Population Health
To find the event types that are most harmful to population health, the number of casualties are aggregated by the event type.
library(plyr)
casualties <- ddply(storm, .(EVTYPE), summarize,
fatalities = sum(FATALITIES),
injuries = sum(INJURIES))
# Find events that caused most death and injury
fatal_events <- head(casualties[order(casualties$fatalities, decreasing = TRUE), ], 10)
injury_events <- head(casualties[order(casualties$injuries, decreasing = TRUE), ], 10)Top 10 events that caused largest number of deaths are
fatal_events[, c("EVTYPE", "fatalities")]## EVTYPE fatalities
## 737 tornado 5633
## 109 excessive heat 1903
## 132 flash flood 978
## 234 heat 937
## 400 lightning 816
## 760 tstm wind 504
## 148 flood 470
## 511 rip current 368
## 309 high wind 248
## 11 avalanche 224Top 10 events that caused most number of injuries are
injury_events[, c("EVTYPE", "injuries")]## EVTYPE injuries
## 737 tornado 91346
## 760 tstm wind 6957
## 148 flood 6789
## 109 excessive heat 6525
## 400 lightning 5230
## 234 heat 2100
## 377 ice storm 1975
## 132 flash flood 1777
## 670 thunderstorm wind 1488
## 203 hail 1361# Economic Effects of Weather Events
To analyze the impact of weather events on the economy, available property damage and crop damage reportings/estimates were used.
In the raw data, the property damage is represented with two fields, a number PROPDMG in dollars and the exponent PROPDMGEXP. Similarly, the crop damage is represented using two fields, CROPDMG and CROPDMGEXP. The first step in the analysis is to calculate the property and crop damage for each event.
exp_transform <- function(e) {
# h -> hundred, k -> thousand, m -> million, b -> billion
if (e %in% c('h', 'H'))
return(2)
else if (e %in% c('k', 'K'))
return(3)
else if (e %in% c('m', 'M'))
return(6)
else if (e %in% c('b', 'B'))
return(9)
else if (!is.na(as.numeric(e))) # if a digit
return(as.numeric(e))
else if (e %in% c('', '-', '?', '+'))
return(0)
else {
stop("Invalid exponent value.")
}
}prop_dmg_exp <- sapply(storm$PROPDMGEXP, FUN=exp_transform)
storm$prop_dmg <- storm$PROPDMG * (10 ** prop_dmg_exp)
crop_dmg_exp <- sapply(storm$CROPDMGEXP, FUN=exp_transform)
storm$crop_dmg <- storm$CROPDMG * (10 ** crop_dmg_exp)# Compute the economic loss by event type
library(plyr)
econ_loss <- ddply(storm, .(EVTYPE), summarize,
prop_dmg = sum(prop_dmg),
crop_dmg = sum(crop_dmg))
# filter out events that caused no economic loss
econ_loss <- econ_loss[(econ_loss$prop_dmg > 0 | econ_loss$crop_dmg > 0), ]
prop_dmg_events <- head(econ_loss[order(econ_loss$prop_dmg, decreasing = T), ], 10)
crop_dmg_events <- head(econ_loss[order(econ_loss$crop_dmg, decreasing = T), ], 10)Top 10 events that caused most property damage (in dollars) are as follows
prop_dmg_events[, c("EVTYPE", "prop_dmg")]## EVTYPE prop_dmg
## 132 flash flood 6.820237e+13
## 694 thunderstorm winds 2.086532e+13
## 737 tornado 1.078951e+12
## 203 hail 3.157558e+11
## 400 lightning 1.729433e+11
## 148 flood 1.446577e+11
## 361 hurricane typhoon 6.930584e+10
## 155 flooding 5.920826e+10
## 581 storm surge 4.332354e+10
## 264 heavy snow 1.793259e+10Similarly, the events that caused biggest crop damage are
crop_dmg_events[, c("EVTYPE", "crop_dmg")]## EVTYPE crop_dmg
## 77 drought 13972566000
## 148 flood 5661968450
## 515 river flood 5029459000
## 377 ice storm 5022113500
## 203 hail 3025974480
## 352 hurricane 2741910000
## 361 hurricane typhoon 2607872800
## 132 flash flood 1421317100
## 118 extreme cold 1312973000
## 179 frost freeze 1094186000## Results
Health impact of weather events
The following plot shows top dangerous weather event types.
library(ggplot2)
library(gridExtra)
# Set the levels in order
p1 <- ggplot(data=fatal_events,
aes(x=reorder(EVTYPE, fatalities), y=fatalities, fill=fatalities)) +
geom_bar(stat="identity") +
coord_flip() +
ylab("Total number of fatalities") +
xlab("Event type") +
theme(legend.position="none") +
ggtitle("Top deadly weather events in the US (1950-2011)")
p2 <- ggplot(data=injury_events,
aes(x=reorder(EVTYPE, injuries), y=injuries, fill=injuries)) +
geom_bar(stat="identity") +
coord_flip() +
ylab("Total number of injuries") +
xlab("Event type") +
theme(legend.position="none") +
ggtitle("Top deadly weather events in the US (1950-2011)")
grid.arrange(p1, p2)
Tornadoes cause most number of deaths and injuries among all event types. There are more than 5,000 deaths and more than 10,000 injuries in the last 60 years in US, due to tornadoes. The other event types that are most dangerous with respect to population health are excessive heat and flash floods.
# 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.
library(ggplot2)
library(gridExtra)
# Set the levels in order
p1 <- ggplot(data=prop_dmg_events,
aes(x=reorder(EVTYPE, prop_dmg), y=log10(prop_dmg), fill=prop_dmg)) +
geom_bar(stat="identity") +
coord_flip() +
xlab("Event type") +
ylab("Property damage in USD (log-scale)") +
theme(legend.position="none") +
ggtitle("Weather costs to the US economy (1950-2011)")
p2 <- ggplot(data=crop_dmg_events,
aes(x=reorder(EVTYPE, crop_dmg), y=crop_dmg, fill=crop_dmg)) +
geom_bar(stat="identity") +
coord_flip() +
xlab("Event type") +
ylab("Crop damage in USD") +
theme(legend.position="none") +
ggtitle("Weather costs to the US economy (1950-2011)")
grid.arrange(p1, p2)
Conclusion
Results reveals tornadoes have significant impact followed by excessive heat under fatalities. Flash floods and thunderstorm winds caused billions of dollars in property damages between 1950 and 2011. The largest crop damage caused was caused by drought, followed by flood, storm and hail.
The data shows that flash floods and thunderstorm winds cost the largest property damages among weather-related natural diseasters.
The most severe weather event in terms of crop damage is the drought. Drought has caused more than 10 billion dollars damage. Other severe crop-damage-causing event types are floods and ice-stroms.