In fmax85/hurricane: Storm observations geom

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

The rationale of this package is to grant peer-grading for the Coursera course of Building Data Visualization Tools. As a potentially useful byproduct, the code in this package contains a customized stat/geom pair named *_hurricane that can be used to visualize storm observation data. Particularly, provided data are arranged in an appropriate way, this geom draws the regions of a storm observation corresponding to the same wind speed level, for the four quadrants defined by an origin point. The contour of the "iso-speed" levels are connected by a polygon grid object.

Introduction

The storm observation data that can be fed to *_hurricane functions are based on the dataset available at Tropical cyclone EBTD. An R script describing the required data frame generating process is provided in the package.

Package content

The package contains the following files and folders:

R -- includes the R scripts of the {stat/geom}_hurricane functions
man -- includes the documentation of the {stat/geom}_hurricane functions, as well as the ike dataset provided along with the package
data -- includes the ike dataset, generated by the data_cleaner.R script
resource -- includes the data_cleaner.R script used to generate the appropriate data frame for the {stat/geom}_hurricane functions, as well as the make_figure.R script adopted to produce the Figure below
vignettes -- includes this vignette
figures -- includes the picture shown below, generated using the geom_hurricane function in the make_figure.R script
NAMESPACE -- NAMESPACE file for this package
DESCRIPTION -- package descritpion file
ike-13sep.png -- picture (same as below) generated using the geom_hurricane function

knitr::include_graphics("../figures/ike-13sep.png")

Code & Data

The reference data of the tropical cyclons is provided in the EBTD database. The ike data frame constitutes a subset of the aforementioned database and is provided along with this package. The format of the data frame is different from the raw data and is strictly required by the {stat/geom}_hurricane functions. For convenience the ike data frame is shown below:

library(magrittr)
library(dplyr)

load(file = "../data/ike.rda")

ike

The ike data frame is generated with the following code (see the data_cleaner.R script, in the resource folder):

# Load and tidy ebrtk data
#
library(magrittr)
library(dplyr)
library(readr)
library(tidyr)
library(stringr)
library(lubridate)
#
# read source data
ext_tracks_widths <- c(7, 10, 2, 2, 3, 5, 5, 6, 4, 5, 4, 4, 5, 3, 4, 3, 3, 3,
                       4, 3, 3, 3, 4, 3, 3, 3, 2, 6, 1)
ext_tracks_colnames <- c("storm_id", "storm_name", "month", "day",
                         "hour", "year", "latitude", "longitude",
                         "max_wind", "min_pressure", "rad_max_wind",
                         "eye_diameter", "pressure_1", "pressure_2",
                         paste("radius_34", c("ne", "se", "sw", "nw"), sep = "_"),
                         paste("radius_50", c("ne", "se", "sw", "nw"), sep = "_"),
                         paste("radius_64", c("ne", "se", "sw", "nw"), sep = "_"),
                         "storm_type", "distance_to_land", "final")
ext_tracks <- read_fwf("ebtrk_atlc_1988_2015.txt",
                       fwf_widths(ext_tracks_widths, ext_tracks_colnames),
                       na = "-99")
#
# refactor quadrant radii
features <- c("storm_name", "year", "month", "day", "hour", "latitude", "longitude")
tmp <- ext_tracks %>% select(features, starts_with("radius_")) %>%
    filter(storm_name == "IKE") %>%
    gather(key, value, -features) %>%
    extract(key, c("wind_speed", "quadrant"), "(\\d+)_(..)") %>%
    spread(quadrant, value)
#
# add (remove) new (old) features
ike <- tmp %>% mutate(date = as.character(make_datetime(as.integer(year), as.integer(month),
                                                        as.integer(day), as.integer(hour)))) %>%
    mutate(longitude = -longitude) %>%
    mutate(storm_id = paste(str_to_title(storm_name), year, sep = "-")) %>%
    select(storm_id, date, latitude, longitude, wind_speed, ne, nw, se, sw)

The stat_hurricane function builds the points, which form the contours of the wind radii, by means of the \link[geosphere]{destPoint} function. geom_hurricane inherits from geom_polygon. The default geom associated to stat_hurricane is geom_hurricane, as shown by the following code chunk:

library(ggplot2)

ike_1 <- ike %>% dplyr::filter(latitude > 29 & latitude < 30)
ggplot(data = ike_1, mapping = aes(x = longitude, y = latitude)) +
    stat_hurricane(mapping = aes(x = longitude, y = latitude, r_ne = ne, r_se = se, r_sw = sw,
                                 r_nw = nw, colour = wind_speed), alpha = 0) +
    theme_minimal()

knitr::include_graphics("../figures/ex_stat_hurricane.png")

Warning

When adding a geom_hurricane layer on top of a map object, as in the first figure, the boundaries of the latitude and longitude of the complete picture map is determined by the map object. In this case, if any points of the wind radii computed by geom_hurricane exceed those limits, the resulting plot will be incorrect. It is advised to use the scale_radii aestethic to reduce the size of the geom_hurricane polygons or to use a less zoomed-in map object.

References

R.D. Peng, S. Kross, and B. Anderson, “Mastering Software Development in R”, section 4.7: Building New Graphical Elements
Documentation of the ggplot2::geom_polygon function, version 2.2.1.9000
Vignette from the ggplot2 package, Extending ggplot2
Vignette from the geosphere package, Introduction to the geosphere package