# SPDS
library(tidyverse)
library(sf)
library(units)

# Data
library(USAboundaries)
library(rnaturalearth)

# Visualization
library(gghighlight)
library(ggrepel)
library(knitr)

Question 1

eqdc = '+proj=eqdc +lat_0=40 +lon_0=-96 +lat_1=20 +lat_2=60 +x_0=0 +y_0=0 +datum=NAD83 +units=m +no_defs'
conus = USAboundaries::us_states() %>% 
  filter(!state_name %in% c("Puerto Rico", "Alaska", "Hawaii")) %>% 
  st_transform(eqdc)
world_boundaries <- rnaturalearth::countries110
world_boundaries <- world_boundaries %>% 
  st_as_sf() %>% 
  filter(admin %in% c("United States of America", "Mexico", "Canada")) %>% 
  st_transform(eqdc)
city_location = readr::read_csv("../data/uscities.csv") %>%
  st_as_sf(coords = c("lng", "lat"), crs = 4326) %>% 
  st_transform(eqdc) %>% 
  filter(!state_name %in% c("Alaska", "Hawaii", "Puerto Rico"))

Question 2

conus_Q2 <- conus %>% 
  st_union() %>% 
  st_cast("MULTILINESTRING")

city_location_Q2<- city_location %>%
  mutate(distance_to_USABorder = st_distance(city_location, conus_Q2), distance_to_USABorder = units::set_units(distance_to_USABorder, "km"), distance_to_USABorder = units::drop_units(distance_to_USABorder))
    
distance_Q2 <- city_location_Q2 %>% 
  select(city, state_name, distance_to_USABorder) %>% 
  arrange(-distance_to_USABorder) %>% 
  slice_head(n = 5) %>% 
  st_drop_geometry() 
knitr::kable(distance_Q2, caption = "Five Cities Farthest from the US Border", col.names = c("City", "State", "Distance (km)"))
Five Cities Farthest from the US Border
City State Distance (km)
Dresden Kansas 1012.317
Herndon Kansas 1007.750
Hill City Kansas 1005.147
Atwood Kansas 1004.734
Jennings Kansas 1003.646 
conus_Q2_2 <- conus %>% 
  st_combine() %>% 
  st_cast("MULTILINESTRING")

city_location_Q2_2 <- city_location %>%
  mutate(distance_to_state = st_distance(city_location, conus_Q2_2), distance_to_state = units::set_units(distance_to_state, "km"), distance_to_state = units::drop_units(distance_to_state))
    
distance_Q2_2 <- city_location_Q2_2 %>% 
  select(city, state_name, distance_to_state) %>% 
  arrange(-distance_to_state) %>% 
  slice_head(n = 5) %>% 
  st_drop_geometry() 
knitr::kable(distance_Q2_2, caption = "Five Cities Farthest from a State Border", col.names = c("City", "State", "Distance (km)"))
Five Cities Farthest from a State Border
City State Distance (km)
Lampasas Texas 308.9216
Bertram Texas 302.8190
Kempner Texas 302.5912
Harker Heights Texas 298.8125
Florence Texas 298.6804 
mexico <-world_boundaries %>% 
  filter(admin == "Mexico") %>% 
  st_cast("MULTILINESTRING")

city_location_Q2_3 <- city_location %>%
  mutate(distance_to_Mexico = st_distance(city_location, mexico), distance_to_Mexico = units::set_units(distance_to_Mexico, "km"), distance_to_Mexico = units::drop_units(distance_to_Mexico))
    
mexico_Q2 <- city_location_Q2_3 %>% 
  select(city, state_name, distance_to_Mexico) %>% 
  arrange(-distance_to_Mexico) %>% 
  slice_head(n = 5) %>% 
  st_drop_geometry() 
knitr::kable(mexico_Q2, caption = "Five Cities Farthest from the Mexican Border", col.names = c("City", "State", "Distance (km)"))
Five Cities Farthest from the Mexican Border
City State Distance (km)
Caribou Maine 3250.334
Presque Isle Maine 3234.570
Calais Maine 3134.348
Eastport Maine 3125.624
Old Town Maine 3048.366 
canada <-world_boundaries %>% 
  filter(admin == "Canada") %>% 
  st_cast("MULTILINESTRING")

city_location_Q2_4 <- city_location %>%
  mutate(distance_to_Canada = st_distance(city_location, canada), distance_to_Canada = units::set_units(distance_to_Canada, "km"), distance_to_Canada = units::drop_units(distance_to_Canada))
    
canada_Q2 <- city_location_Q2_4 %>% 
  select(city, state_name, distance_to_Canada) %>% 
  arrange(-distance_to_Canada) %>% 
  slice_head(n = 5) %>% 
  st_drop_geometry() 
knitr::kable(canada_Q2, caption = "Five Cities Farthest from the Canadian Border", col.names = c("City", "State", "Distance (km)"))
Five Cities Farthest from the Canadian Border
City State Distance (km)
Guadalupe Guerra Texas 2206.455
Sandoval Texas 2205.641
Fronton Texas 2204.784
Fronton Ranchettes Texas 2202.118
Evergreen Texas 2202.020

Question 3

city_location_Q3 <- city_location %>% 
  slice_max(population, n = 10)
ggplot() +
  geom_sf(data = world_boundaries, color = "black") +
  geom_sf(data = conus, color = "#F5B8B5") +
  geom_sf(data = city_location_Q3, color = "red") +
  ggrepel::geom_label_repel(data = city_location_Q3, aes(label = city, geometry = geometry), stat = "sf_coordinates")+
  labs(title = "10 Largest USA Cities (by Population)",
       x = "Longitude",
       y = "Latitude")

distance_Q3 <- city_location_Q2 %>% 
  slice_max(distance_to_USABorder, n = 5) 
ggplot() +
  geom_sf(data = conus) +
  geom_sf(data = city_location_Q2, aes(col = distance_to_USABorder)) +
  geom_sf(data = distance_Q3)+
  scale_color_gradient(low = 'red', high = 'black') +
  ggthemes::theme_map() +
  ggrepel::geom_label_repel(data = distance_Q3, aes(label = city, geometry = geometry), stat = "sf_coordinates")+
  labs(title = "City Distance from the Border", color = "Distance (km)")

distance_Q3_3 <- city_location_Q2_2 %>% 
  slice_max(distance_to_state, n = 5)
  
ggplot() +
  geom_sf(data = conus) +
  geom_sf(data = city_location_Q2_2, aes(col = distance_to_state)) +
  geom_sf(data = distance_Q3_3)+
  scale_color_gradient(low = 'red', high = 'black') +
  ggthemes::theme_map() +
  ggrepel::geom_label_repel(data = distance_Q3_3, aes(label = city, geometry = geometry), stat = "sf_coordinates")+
  labs(title = "City Distance from Nearest State", color = "Distance (km)")

equidistance <- city_location %>%
  mutate(distance_to_Mexico = st_distance(city_location, mexico), distance_to_Mexico = units::set_units(distance_to_Mexico, "km"), distance_to_Mexico = units::drop_units(distance_to_Mexico)) %>% 
  mutate(distance_to_Canada = st_distance(city_location, canada), distance_to_Canada = units::set_units(distance_to_Canada, "km"), distance_to_Canada = units::drop_units(distance_to_Canada)) %>% 
  mutate(diff = abs(distance_to_Canada - distance_to_Mexico)) %>% 
  mutate(check = 100)

equidistance_Q3 <- equidistance %>% 
  filter(diff < 100) %>% 
  slice_max(population, n = 5)

ggplot()+
  geom_sf(data = conus) +
  geom_sf(data = equidistance, aes(color = diff)) +
  geom_sf(data = equidistance_Q3, color = "red") +
  scale_color_gradient(low = "blue", high = "black") +
  ggthemes::theme_map() +
  ggrepel::geom_label_repel(data = equidistance_Q3, aes(label = city, geometry = geometry), stat = "sf_coordinates")+
  labs(title = " Equidistance boundary from Mexico and Canada")

Question 4

border_Q4 <- city_location_Q2 %>%
  mutate(totalpopulation = sum(population)) %>%
  filter(distance_to_USABorder < 160) %>% 
  summarize(cities = n(), ppl = sum(population), percentage = (ppl / totalpopulation) * 100) %>% 
  slice(n = 1) %>% 
  st_drop_geometry()
knitr::kable(border_Q4, caption = "Quantifing Border Zone", col.names = c("Number of Cities", "Population within 100 Mile Zone", "Percentage of Total Population in 100 Mile Zone"), format.args = list(big.mark = ","))
Quantifing Border Zone
Number of Cities Population within 100 Mile Zone Percentage of Total Population in 100 Mile Zone
12,283 259,935,815 65.43979

The data match the ACLU estimate in the link that roughly two-thirds of the United States’ population lives within the 100-mile zone. That’s about 200 million people 

border <- city_location_Q2 %>% 
  filter(distance_to_USABorder < 160)

border_zone <- city_location_Q2%>% 
  filter(distance_to_USABorder < 160) %>% 
  group_by(state_name) %>% 
  slice_max(population, n = 1)

ggplot()+
  geom_sf(data = conus, fill = NA) +
  geom_sf(data = city_location_Q2, aes(color = distance_to_USABorder)) +
  geom_sf(data = border_zone, color = "black") +
  scale_color_gradient(low = "orange", high = "darkred") +
  gghighlight(distance_to_USABorder < 160) +
  ggthemes::theme_map() +
  ggrepel::geom_label_repel(data = border_zone, aes(label = city, geometry = geometry), stat = "sf_coordinates") +
  labs(title = "Most populous city in each state within the Danger Zone")