Tornado causes the most loss of life, flood causes the most loss of money
Abidur Rahman
2024-04-01
Synopsis
Data from the NOAA Storm Database was analyzed to figure out the types of events that cause the most loss of life and money. The data was transformed to sum up the losses from each type of events. Then they were graphed to show the effects visually. From this we can see that Tornado causes the most loss of life, while Flood causes the most loss of money.
Data processing
Loading the libraries
library(readr)## Warning: package 'readr' was built under R version 4.3.3library(dplyr)## Warning: package 'dplyr' was built under R version 4.3.3##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(ggplot2)## Warning: package 'ggplot2' was built under R version 4.3.3Loading the data
and transforming the character variables to factors
The function read_csv from readr package was used to directly read
the compressed .bz2 file.
Then all the chr type columns were transformed into factors for easier
analysis.
Storm.Data <- read_csv("repdata_data_StormData.csv.bz2") ## Rows: 902297 Columns: 37
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (18): BGN_DATE, BGN_TIME, TIME_ZONE, COUNTYNAME, STATE, EVTYPE, BGN_AZI,...
## dbl (18): STATE__, COUNTY, BGN_RANGE, COUNTY_END, END_RANGE, LENGTH, WIDTH, ...
## lgl (1): COUNTYENDN
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.Storm.Data <- Storm.Data %>% mutate_if(is.character, factor)Calculating the total fatality and injury for each type of events
After grouping by EVTYPE, sum of all fatalities and injuries were
calculated using the summarize function.
Then the data frame is sorted in descending order of Total fatality,
then total injury.
Population.Health <- Storm.Data %>%
group_by(EVTYPE) %>%
summarize(Total.fatality = sum(FATALITIES), Total.injury = sum(INJURIES)) %>%
arrange(desc(Total.fatality), desc(Total.injury))Calculating the total damage expenditure for each type of events
Both property damage and crop damage data were saved using two
columns each. One for the value, and one for the suffix to indicate the
exponentiaion.
First these had to be combined to form a single numerical column each,
for property damage and crop damage.
After grouping by EVTYPE, sum of all property damage and crop damage
were calculated using the summarize function. Then these two were added
to figure out the Total damage for each type.
Then the data frame is sorted in descending order of Total damage, then
Total property damage, then Total crop damage.
Economic.Damage <- Storm.Data %>%
mutate(
PROPDMG1 = case_when(
.default = PROPDMG,
PROPDMGEXP == "B" ~ PROPDMG*1000000000,
PROPDMGEXP == "M" ~ PROPDMG*1000000,
PROPDMGEXP == "m" ~ PROPDMG*1000000,
PROPDMGEXP == "K" ~ PROPDMG*1000,
PROPDMGEXP == "k" ~ PROPDMG*1000,
PROPDMGEXP == "H" ~ PROPDMG*100,
PROPDMGEXP == "h" ~ PROPDMG*100,
PROPDMGEXP == "8" ~ PROPDMG*100000000,
PROPDMGEXP == "7" ~ PROPDMG*10000000,
PROPDMGEXP == "6" ~ PROPDMG*1000000,
PROPDMGEXP == "5" ~ PROPDMG*100000,
PROPDMGEXP == "4" ~ PROPDMG*10000,
PROPDMGEXP == "3" ~ PROPDMG*1000,
PROPDMGEXP == "2" ~ PROPDMG*100,
PROPDMGEXP == "1" ~ PROPDMG*10
),
CROPDMG1 = case_when(
.default = CROPDMG,
CROPDMGEXP == "B" ~ CROPDMG*1000000000,
CROPDMGEXP == "M" ~ CROPDMG*1000000,
CROPDMGEXP == "m" ~ CROPDMG*1000000,
CROPDMGEXP == "K" ~ CROPDMG*1000,
CROPDMGEXP == "k" ~ CROPDMG*1000,
CROPDMGEXP == "H" ~ CROPDMG*100,
CROPDMGEXP == "h" ~ CROPDMG*100,
CROPDMGEXP == "8" ~ CROPDMG*100000000,
CROPDMGEXP == "7" ~ CROPDMG*10000000,
CROPDMGEXP == "6" ~ CROPDMG*1000000,
CROPDMGEXP == "5" ~ CROPDMG*100000,
CROPDMGEXP == "4" ~ CROPDMG*10000,
CROPDMGEXP == "3" ~ CROPDMG*1000,
CROPDMGEXP == "2" ~ CROPDMG*100,
CROPDMGEXP == "1" ~ CROPDMG*10
)
) %>%
group_by(EVTYPE) %>%
summarize(Total.PROPDMG = sum(PROPDMG1), Total.CROPDMG = sum(CROPDMG1)) %>%
mutate(Total.DMG = Total.PROPDMG + Total.CROPDMG) %>%
arrange(desc(Total.DMG), desc(Total.PROPDMG), desc(Total.CROPDMG))Analysis
Top 10 types of events causing most fatality and injury
Population.Health[1:10,]## # A tibble: 10 × 3
## EVTYPE Total.fatality Total.injury
## <fct> <dbl> <dbl>
## 1 TORNADO 5633 91346
## 2 EXCESSIVE HEAT 1903 6525
## 3 FLASH FLOOD 978 1777
## 4 HEAT 937 2100
## 5 LIGHTNING 816 5230
## 6 TSTM WIND 504 6957
## 7 FLOOD 470 6789
## 8 RIP CURRENT 368 232
## 9 HIGH WIND 248 1137
## 10 AVALANCHE 224 170Let’s see using graphs how significant these are
I deliberately choose to not equate fatalities with specific numbers of injuries and thus plot both in a scatter plot.
ggplot(Population.Health, aes(Total.fatality, Total.injury)) +
geom_jitter()
The first data, ie, Tornado far surpasses everythis else, so it is not that clear how many of the other types are doing the most damage,
Using log10 scale for better visualization
ggplot(Population.Health, aes(log10(Total.fatality), log10(Total.injury))) +
geom_jitter()
It appears that the first seven of the types are the ones most responsible here, even in log scale.
Top 10 types of events causing most expenditure
Economic.Damage[1:10,]## # A tibble: 10 × 4
## EVTYPE Total.PROPDMG Total.CROPDMG Total.DMG
## <fct> <dbl> <dbl> <dbl>
## 1 FLOOD 144657709807 5661968450 150319678257
## 2 HURRICANE/TYPHOON 69305840000 2607872800 71913712800
## 3 TORNADO 56947380676. 414953270 57362333946.
## 4 STORM SURGE 43323536000 5000 43323541000
## 5 HAIL 15735267513. 3025954473 18761221986.
## 6 FLASH FLOOD 16822723978. 1421317100 18244041078.
## 7 DROUGHT 1046106000 13972566000 15018672000
## 8 HURRICANE 11868319010 2741910000 14610229010
## 9 RIVER FLOOD 5118945500 5029459000 10148404500
## 10 ICE STORM 3944927860 5022113500 8967041360Let’s see using a graph how significant these are
ggplot(Economic.Damage[1:10,], aes(EVTYPE, Total.DMG)) +
geom_col() +
guides(x = guide_axis(angle = 90))
It appears that the first four of the types are the ones most responsible here.
Results
Types of events most harmful with respect to population health:
with total fatalities and total injury by each
Population.Health[1:7,]## # A tibble: 7 × 3
## EVTYPE Total.fatality Total.injury
## <fct> <dbl> <dbl>
## 1 TORNADO 5633 91346
## 2 EXCESSIVE HEAT 1903 6525
## 3 FLASH FLOOD 978 1777
## 4 HEAT 937 2100
## 5 LIGHTNING 816 5230
## 6 TSTM WIND 504 6957
## 7 FLOOD 470 6789Types of events having the greatest economic consequences:
with total damage, total property damage and total crop damage caused by each.
Economic.Damage[1:4, c(1, 4, 2, 3)]## # A tibble: 4 × 4
## EVTYPE Total.DMG Total.PROPDMG Total.CROPDMG
## <fct> <dbl> <dbl> <dbl>
## 1 FLOOD 150319678257 144657709807 5661968450
## 2 HURRICANE/TYPHOON 71913712800 69305840000 2607872800
## 3 TORNADO 57362333946. 56947380676. 414953270
## 4 STORM SURGE 43323541000 43323536000 5000