library(tidyverse)
library(sf)
library(units)
library(ggplot2)
library(ggrepel)
library(gghighlight)
library(ggthemes)
library(knitr)

region = data.frame(region = state.region, 
                    state_name = state.name)

south = right_join(USAboundaries::us_states(),
                   region,
                   by = "state_name") %>% 
  filter(region == "South")

plot(south['aland'])

cities = readr::read_csv("~/github/geog-176A-labs/data/uscities.csv") %>% 
  st_as_sf(coords = c("lng", "lat"), crs = 4326) %>% 
  st_filter(south, .predicate = st_intersects)

plot(south$geometry)
plot(cities, add= TRUE, pch = 16, cex = .1)

south_c = st_combine(south) %>% 
  st_cast("MULTILINESTRING")

south_c = st_transform(south_c, 5070)
cities = st_transform(cities, 5070)

cities = cities %>% 
  mutate(dist_to_state = st_distance(cities, south_c),
         dist_to_state = units::set_units(dist_to_state, "km"),
         dist_to_state = units::drop_units(dist_to_state))

big_cities = cities %>%
  group_by(state_name) %>%
  slice_max(population, n = 2)



ggplot() +
  geom_sf(data = south_c) +
  geom_sf(data = cities, aes(col = dist_to_state), size = .1) +
  geom_sf(data = big_cities, col = "navy") +
  scale_color_gradient(low = "gray", high = "red") +
  ggthemes::theme_map() +
  ggrepel::geom_label_repel(
    data = big_cities,
    aes(label = city, geometry = geometry),
    stat = "sf_coordinates",
    size = 4
  ) +
  labs(title = "Labeling Example",
       col = "Distance (km)")

ggplot() +
  geom_sf(data = south_c) +
  geom_sf(data = cities, aes(col = dist_to_state), size = .1) +
  gghighlight::gghighlight(population > 10000) +
  geom_sf(data = big_cities, col = "navy") +
  scale_color_gradient(low = "gray", high = "red") +
  ggthemes::theme_map() +
  labs(title = "Highlighting Example",
       col = "Distance (km)")

#Question 1:

###1.1

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'

###1.2

library(USAboundaries) 

region = data.frame(region = state.region, 
                    state_name = state.name)


USA_states = right_join(USAboundaries::us_states(),
                   region,
                   by = "state_name") %>% 
  filter(!state_abbr %in% c("PR", "AK", "HI")) 

USA_states <- st_transform(USA_states, '+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')
  
plot(USA_states$geometry, main = "Map of Continental United States ")

#1.3

library(rnaturalearth)

boundaries = rnaturalearth::countries110 %>% 
  st_as_sf(crs = 4269) %>% 
  filter(SOVEREIGNT %in% c ("United States of America", "Canada", "Mexico")) %>% 
  st_transform(eqdc)

plot(boundaries$geometry)

#1.4

library(readr)
cities = readr::read_csv("~/github/geog-176A-labs/data/uscities.csv") %>% 
  st_as_sf(coords = c("lng", "lat"), crs = 4326) %>% 
  filter(!state_name %in% c("Hawaii", "Alaska", "Puerto Rico")) %>%
  st_transform(eqdc)

#Quesrion 2: #2.1

USA_states_string = st_union(USA_states) %>% 
  st_cast("MULTILINESTRING")

cities = cities %>% 
  mutate(dist2border = st_distance(cities, USA_states_string),
         dist2border = units::set_units(dist2border, "km"),
         dist2border = units::drop_units(dist2border))

furthest_cities = cities %>%
  slice_max(dist2border, n = 5) %>% 
  select(city, state_name, dist2border) %>% 
  st_drop_geometry()

kable(furthest_cities, caption = "The Furthest Cities From the American Boarders",
      col.names = c("City", "State", "Distance to Border in km"),
      format.args = list(big.mark = ","))

The Furthest Cities From the American Boarders

City	State	Distance to Border in km
Dresden	Kansas	1,012.317
Herndon	Kansas	1,007.750
Hill City	Kansas	1,005.147
Atwood	Kansas	1,004.734
Jennings	Kansas	1,003.646

#2.2

USA_states_string2 = st_combine(USA_states) %>% 
  st_cast("MULTILINESTRING")

cities = cities %>% 
  mutate(dist2state = st_distance(cities, USA_states_string2),
         dist2state = units::set_units(dist2state, "km"),
         dist2state = units::drop_units(dist2state))

furthest_cities_from_state = cities %>% 
  slice_max(dist2state, n = 5) %>% 
  select(city, state_name, dist2state) %>% 
  st_drop_geometry()

kable(furthest_cities_from_state, caption = "The Furthest Cities From State Border",
      col.names = c("City", "State", "Distance to State Border in km"),
      format.args = list(big.mark = ","))

The Furthest Cities From State Border

City	State	Distance to State Border in km
Lampasas	Texas	308.9216
Bertram	Texas	302.8190
Kempner	Texas	302.5912
Harker Heights	Texas	298.8125
Florence	Texas	298.6804

#2.3

Mexico_border = boundaries %>% 
filter(SOVEREIGNT == "Mexico")

cities = cities %>% 
  mutate(dist2Mexico = st_distance(cities, Mexico_border),
         dist2Mexico = units::set_units(dist2Mexico, "km"),
         dist2Mexico = units::drop_units(dist2Mexico))

furthest_cities_from_Mexico = cities %>% 
  slice_max(dist2Mexico, n = 5) %>% 
  select(city, state_name, dist2Mexico) %>% 
  st_drop_geometry()

kable(furthest_cities_from_Mexico, caption = "Furthest Cities From Mexico Border",
      col.names = c("City", "State", "Distance to Mexico Border in km"),
      format.args = list(big.mark = ","))

Furthest Cities From Mexico Border

City	State	Distance to Mexico Border in km
Caribou	Maine	3,250.334
Presque Isle	Maine	3,234.570
Calais	Maine	3,134.348
Eastport	Maine	3,125.624
Old Town	Maine	3,048.366

#2.4

canada_border = boundaries %>% 
filter(SOVEREIGNT == "Canada")

cities = cities %>% 
  mutate(dist2Canada = st_distance(cities, canada_border),
         dist2Canada = units::set_units(dist2Canada, "km"),
         dist2Canada = units::drop_units(dist2Canada))

furthest_cities_from_Canada = cities %>% 
  slice_max(dist2Canada, n = 5) %>% 
  select(city, state_name, dist2Canada) %>% 
  st_drop_geometry()

kable(furthest_cities_from_Canada, caption = "Furthest Cities From Mexico Border",
      col.names = c("City", "State", "Distance to Mexico Border in km"),
      format.args = list(big.mark = ","))

Furthest Cities From Mexico Border

City	State	Distance to Mexico Border in km
Guadalupe Guerra	Texas	2,206.455
Sandoval	Texas	2,205.641
Fronton	Texas	2,204.784
Fronton Ranchettes	Texas	2,202.118
Evergreen	Texas	2,202.020

#Question 3 #3.1

big_cities = cities %>% 
  slice_max(population, n = 10)

ggplot()+
  geom_sf(data = boundaries)+
  geom_sf(data = USA_states_string2)+
  geom_sf(data = big_cities, col = "red", size = 1.5)+
  ggthemes::theme_map()+
  labs(title = "The 10 Most Populated Cities in the US")+
  ggrepel::geom_label_repel(
    data = big_cities,
    aes(label = city, geometry = geometry),
    stat = "sf_coordinates",
    size = 3
  )

#3.2

furthest_cities2 = cities %>% 
  slice_max(dist2border, n = 5) %>% 
  select(city, state_name, dist2border)

ggplot()+
  geom_sf(data = cities, aes(col = dist2border), size = .5)+
  geom_sf(data = furthest_cities2, col = "blue")+
  geom_sf(data = USA_states_string)+
  scale_color_gradient(low = "white", high = "red")+
  ggthemes::theme_map()+
  labs(title = "5 Furthest Cities From Country Borders and Distance From Boarder")+
  ggrepel::geom_label_repel(
    data = furthest_cities2,
    aes(label = city, geometry = geometry),
    stat = "sf_coordinates",
    size = 3
  )

#3.3

furthest_cities3 = cities %>% 
  slice_max(dist2state, n = 5) %>% 
  select(city, state_name, dist2state)

ggplot()+
  geom_sf(data = cities, aes(col = dist2state), size = .5)+
  geom_sf(data = furthest_cities3, col = "blue")+
  geom_sf(data = USA_states_string2)+
  scale_color_gradient(low = "white", high = "red")+
  ggthemes::theme_map()+
  labs(title = "5 Furthest Cities From State Borders and Distance From Boarder")+
  ggrepel::geom_label_repel(
    data = furthest_cities3,
    aes(label = city, geometry = geometry),
    stat = "sf_coordinates",
    size = 3
  )

#3.4

cities = cities %>% 
  mutate(distance_Can_Mex = abs(dist2Canada - dist2Mexico))

equa_Can_Mex = cities %>% 
  filter(distance_Can_Mex < 100)

biggest_equa_Can_Mex = equa_Can_Mex %>% 
  slice_max(population, n = 5)

ggplot()+
  geom_sf(data = boundaries)+
  geom_sf(data = USA_states_string)+
  geom_sf(data = equa_Can_Mex, color = "green", size = .5)+
  geom_sf(data = biggest_equa_Can_Mex, color = "red", size = .5)+
  ggthemes::theme_map()+
  labs(title = "Cities Equadistant from Canada and Mexico Borders")+
  ggrepel::geom_label_repel(
    data = biggest_equa_Can_Mex,
    aes(label = city, geometry = geometry),
    stat = "sf_coordinates",
    size = 3
  )

#Question 4 #4.1

Border_zone = cities %>% 
  mutate(Mex_cities = dist2Mexico < 160) %>% 
  mutate(Can_cities = dist2Canada < 160) %>%
  filter(!Mex_cities %in% "FALSE" | !Can_cities %in% "FALSE")

#According to the data. there are 3296 cities in the boarder zone. 

sum(Border_zone$population)

## [1] 47082122

#According to the sum function for Border_zone, the population in the border zone is 47,082,122. This however may be skewed becauset the website says there shoul be 200 million people in the border zone. 

sum(cities$population)

## [1] 397213686

#According to the sum function for the cities data frame, there are 397,216,686 people in the United States. This means. That the ratio of people in the border zone to people in the US is as follows:

sum(Border_zone$population) / sum(cities$population)

## [1] 0.118531

#4.2
ggplot()+
  geom_sf(data = cities, aes(col = dist2border), size = .5)+
  geom_sf(data = Border_zone, col = "blue")+
  geom_sf(data = USA_states_string2)+
  scale_color_gradient(low = "white", high = "red")+
  ggthemes::theme_map()+
  labs(title = "5 Furthest Cities From State Borders and Distance From Boarder")+
  ggrepel::geom_label_repel(
    data = furthest_cities3,
    aes(label = city, geometry = geometry),
    stat = "sf_coordinates",
    size = 3
  )