Lines

Decline in Brazil’s railway network

Author

Ana Luisa Bodevan

Day 2: Lines

Challenge Classic: Map linear features (e.g., roads, rivers, migration paths, flow lines). Explore line thickness, color, and direction to convey information.

Brazil’s railway network has declined since its peak in the 20th century because of a number of reasons, from the crisis in coffee market during the Great Depression to a policy of highway expansion and lobby from car manufacturers in the 1950s. BBC has a pretty cool article about it (in portuguese). Overall, Brazilian railway network lost over 8.000 kilometers since the 20th century.

For Day 2 of the #30DayMapChallenge my idea was to map the decline in Brazil’s railway network.

1. Setup

1.1 Load libraries

Code

library(pacman)
pacman :: p_load(tidyverse, sf, here, geobr, rnaturalearth, readxl, ragg, showtext, glue, patchwork, scales, cowplot)

1.2 Load data

The data used is from the Banco de Informações de Transportes (BIT) from the Ministry of Transport. Data of the extension of railways over the years is from IBGE

Code

knitr::opts_knit$set(root.dir = here::here())

# read railway shapefile 
zip_path <- here("2025/02_lines/BaseFerro.zip")
extract_dir <- here("2025/02_lines/BaseFerro")
unzip(zip_path, exdir = extract_dir)
shp_file <- list.files(extract_dir, pattern = "\\.shp$", recursive = TRUE, full.names = TRUE)[1]
shp_rail <- sf::st_read(shp_file, quiet = TRUE)

# get brazil and south america shapefiles 
shp_br <- read_state()

shp_latam <- ne_countries(type = "countries", continent = "south america", returnclass = "sf")

shp_latam <- shp_latam %>% 
  st_transform(crs = st_crs(shp_br))

# read kilometer extension
km <- readxl::read_xls("kms.xls")

2. Data Wrangling

2.1 Tidy kilometer extension data

Code

colnames(km) <- km[3, ]
km <- km[-(1:3), ]
km <- km[-(133:135), ]
rownames(km) <- NULL
colnames(km) <- c("YEAR", "KM")
km$YEAR <- gsub("\\.+", "", km$YEAR) # get the right col names and clean it up 

km$YEAR <- as.integer(km$YEAR)
km$KM <- as.integer(km$KM)

km <- km |> 
  arrange(YEAR) |> 
  mutate(
    delta_km = KM - lag(KM),
    pct_change = 100 * delta_km / lag(KM)
  )

km <- km   |> 
  add_row(YEAR = 2025, KM = 29165) |> 
  arrange(YEAR) |> 
  mutate(
    delta_km = KM - lag(KM),
    pct_change = 100 * delta_km / lag(KM)
  ) # 2025 data from ANTT

2.2 Tidy names in shp_rail

Code

shp_rail <- shp_rail |> 
  mutate(
    status = factor(
      tip_situac, levels = c("Em Operação", "Em Obra", "Planejada", "Estudo", "Desativada", NA),
      labels = c("Operating", "Under Repair", "Under Study", "Under Study", "Deactivated")
    )
  ) |> 
  filter(!is.na(status))

2.3 Create bounding box

We have to exclude the state of Espiríto Santo of the bounding box because it has islands that may distort the map if included.

Code

bbox <- shp_br |> 
  filter(abbrev_state != "ES") |> 
  st_bbox()

3. Plot

3.1 Fonts and text

Code

font_add_google("Inter", "inter", bold.wt = 800, regular.wt = 400)
showtext_auto()

title <- "Brazil's Railways (2023)"
subtitle <- "Since its peak in 1960 with 34,252 km, Brazil's\nrailways have declined by over 8,000 km"
caption <- "#30DayMapChallenge Day 2 - Lines\nData: ANNT & IBGE; Graphic by Ana Bodevan"

3.2 Plot Map

Code

map <- ggplot() +
  # South America context (light background)
  geom_sf(data = shp_latam, fill = "#F5f5f5", color = "gray80", size = 0.15) +
  
  # Brazil states
  geom_sf(data = shp_br, color = "#BDBDBD", fill = "#dfe2e7ff", size = 0.25) +

  # Railways by status
  geom_sf(data = shp_rail, aes(color = status, linetype = status), size = 0.5) +
  
  # Color scale for railways
  scale_color_manual(
    values = c(
      "Operating" = "#2C3E50",           
      "Under Repair" = "#E67E22",        
      "Under Study" = "#4987c2ff",         
      "Deactivated" = "#C0392B"          
    ),
    name = "Railway Status"
  ) +
  
  # Line type according to status 
  scale_linetype_manual(
    values = c(
      "Operating" = "solid",             
      "Under Repair" = "dashed",         
      "Under Study" = "dotted",        
      "Deactivated" = "dotdash"          
    ),
    name = "Railway Status"
  ) +

  # Set bounding box to focus on Brazil
  coord_sf(xlim = c(bbox["xmin"], bbox["xmax"]),
           ylim = c(bbox["ymin"], bbox["ymax"])) +

  # Labs
  labs(tag = "",
      title = title,
      subtitle = subtitle,
      caption = caption) +

  
  # Theme 
  ggthemes::theme_map(base_family = "inter")  %+replace%
  theme(
    plot.background = element_rect(fill = "white", color = NA),
    panel.background = element_rect(fill = "white", color = NA),

    plot.title = element_text(family = "inter", face = "bold", size = 24, 
                              hjust = 0, margin = margin(b = 5)),
    plot.subtitle = element_text(family = "inter", size = 12, 
                                 hjust = 0, margin = margin(b = 15)),
    plot.caption = element_text(family = "inter", size = 8, hjust = 0, color = "gray50"),

    legend.position = c(1.03, 0.35),   # (x, y): 1.0 = right edge, >1 = outside
    legend.justification = c("left", "top"),
    legend.box.margin = margin(0, 0, 0, 0),
    legend.background = element_rect(fill = "white", color = NA),
    legend.title = element_text(face = "bold", size = 11),
    legend.text = element_text(size = 10),

    plot.margin = margin(10, 200, 5, 10)
  )

map

3.3 Plot Line Chart

Code

line <- ggplot(km, aes(x = YEAR, y = KM)) +
  geom_line(color = "#2C3E50", linewidth = 0.8) +

  # Highlight the peak
  geom_vline(xintercept = 1960, linetype = "dashed", color = "#C0392B", alpha = 0.5) +

  scale_y_continuous(
    labels = scales::label_number(scale = 1/1000, suffix = "k"),
    breaks = seq(0, 35000, 5000)
  ) +
  scale_x_continuous(breaks = seq(1854, 2023, 30)) +
  labs(
    x = "Year",
    y = "Railway Extension (km)",
    title = NULL
  ) +
  theme_minimal() +
  theme(
    plot.background = element_rect(fill = "white", color = NA),
    panel.background = element_rect(fill = "white", color = NA),
    panel.grid.minor = element_blank(),
    panel.grid.major = element_line(color = "gray90", size = 0.3),
    axis.title = element_text(family = "inter", size = 10, face = "bold"),
    axis.text = element_text(family = "inter", size = 9, angle = 45),
    axis.title.y = element_text(angle = 90, margin = margin(r = 10)),
    axis.title.x = element_text(margin = margin(t = 10)),
    plot.margin = margin(20, 20, 20, 20)
  )

line

Code

# Add a light border to your line chart for clarity
line_box <- line +
  theme(
    plot.background = element_rect(fill = "white", color = "gray85", linewidth = 0.3),
    panel.grid.major = element_line(color = "gray90", linewidth = 0.25),
    axis.text = element_text(size = 8, family = "inter"),
    axis.title = element_text(size = 9, family = "inter", face = "bold")
  )

final_plot <- ggdraw() +
  draw_plot(map, x = 0, y = 0, width = 1, height = 1) +
  draw_plot(
    line_box,
    x = 0.68,   
    y = 0.35,   
    width = 0.28, 
    height = 0.35,
    hjust = 0,
    vjust = 0
  )

final_plot

--- title: "Lines" author: "Ana Luisa Bodevan" description: "Decline in Brazil's railway network" image: "02_lines.png" execute: warning: false message: false eval: true format: html: code-tools: true code-fold: true --- # Day 2: Lines | Challenge Classic: Map linear features (e.g., roads, rivers, migration paths, flow lines). Explore line thickness, color, and direction to convey information. Brazil's railway network has declined since its peak in the 20th century because of a number of reasons, from the crisis in coffee market during the Great Depression to a policy of highway expansion and lobby from car manufacturers in the 1950s. [BBC](https://www.bbc.com/portuguese/articles/cp3qlyk3vy2o) has a pretty cool article about it (in portuguese). Overall, Brazilian railway network lost over 8.000 kilometers since the 20th century. For Day 2 of the #30DayMapChallenge my idea was to map the decline in Brazil's railway network. ## 1. Setup ### 1.1 Load libraries ```{r} library(pacman) pacman :: p_load(tidyverse, sf, here, geobr, rnaturalearth, readxl, ragg, showtext, glue, patchwork, scales, cowplot) ``` ### 1.2 Load data The data used is from the [Banco de Informações de Transportes](https://www.gov.br/transportes/pt-br/assuntos/dados-de-transportes/bit/bit-mapas) (BIT) from the Ministry of Transport. Data of the extension of railways over the years is from IBGE ```{r, message=FALSE, warning=FALSE} knitr::opts_knit$set(root.dir = here::here()) # read railway shapefile zip_path <- here("2025/02_lines/BaseFerro.zip") extract_dir <- here("2025/02_lines/BaseFerro") unzip(zip_path, exdir = extract_dir) shp_file <- list.files(extract_dir, pattern = "\\.shp$", recursive = TRUE, full.names = TRUE)[1] shp_rail <- sf::st_read(shp_file, quiet = TRUE) # get brazil and south america shapefiles shp_br <- read_state() shp_latam <- ne_countries(type = "countries", continent = "south america", returnclass = "sf") shp_latam <- shp_latam %>% st_transform(crs = st_crs(shp_br)) # read kilometer extension km <- readxl::read_xls("kms.xls") ``` ## 2. Data Wrangling ### 2.1 Tidy kilometer extension data ```{r} colnames(km) <- km[3, ] km <- km[-(1:3), ] km <- km[-(133:135), ] rownames(km) <- NULL colnames(km) <- c("YEAR", "KM") km$YEAR <- gsub("\\.+", "", km$YEAR) # get the right col names and clean it up km$YEAR <- as.integer(km$YEAR) km$KM <- as.integer(km$KM) km <- km |> arrange(YEAR) |> mutate( delta_km = KM - lag(KM), pct_change = 100 * delta_km / lag(KM) ) km <- km |> add_row(YEAR = 2025, KM = 29165) |> arrange(YEAR) |> mutate( delta_km = KM - lag(KM), pct_change = 100 * delta_km / lag(KM) ) # 2025 data from ANTT ``` ### 2.2 Tidy names in shp_rail ```{r} shp_rail <- shp_rail |> mutate( status = factor( tip_situac, levels = c("Em Operação", "Em Obra", "Planejada", "Estudo", "Desativada", NA), labels = c("Operating", "Under Repair", "Under Study", "Under Study", "Deactivated") ) ) |> filter(!is.na(status)) ``` ### 2.3 Create bounding box We have to exclude the state of Espiríto Santo of the bounding box because it has islands that may distort the map if included. ```{r} bbox <- shp_br |> filter(abbrev_state != "ES") |> st_bbox() ``` ## 3. Plot ### 3.1 Fonts and text ```{r} font_add_google("Inter", "inter", bold.wt = 800, regular.wt = 400) showtext_auto() title <- "Brazil's Railways (2023)" subtitle <- "Since its peak in 1960 with 34,252 km, Brazil's\nrailways have declined by over 8,000 km" caption <- "#30DayMapChallenge Day 2 - Lines\nData: ANNT & IBGE; Graphic by Ana Bodevan" ``` ### 3.2 Plot Map ```{r} map <- ggplot() + # South America context (light background) geom_sf(data = shp_latam, fill = "#F5f5f5", color = "gray80", size = 0.15) + # Brazil states geom_sf(data = shp_br, color = "#BDBDBD", fill = "#dfe2e7ff", size = 0.25) + # Railways by status geom_sf(data = shp_rail, aes(color = status, linetype = status), size = 0.5) + # Color scale for railways scale_color_manual( values = c( "Operating" = "#2C3E50", "Under Repair" = "#E67E22", "Under Study" = "#4987c2ff", "Deactivated" = "#C0392B" ), name = "Railway Status" ) + # Line type according to status scale_linetype_manual( values = c( "Operating" = "solid", "Under Repair" = "dashed", "Under Study" = "dotted", "Deactivated" = "dotdash" ), name = "Railway Status" ) + # Set bounding box to focus on Brazil coord_sf(xlim = c(bbox["xmin"], bbox["xmax"]), ylim = c(bbox["ymin"], bbox["ymax"])) + # Labs labs(tag = "", title = title, subtitle = subtitle, caption = caption) + # Theme ggthemes::theme_map(base_family = "inter") %+replace% theme( plot.background = element_rect(fill = "white", color = NA), panel.background = element_rect(fill = "white", color = NA), plot.title = element_text(family = "inter", face = "bold", size = 24, hjust = 0, margin = margin(b = 5)), plot.subtitle = element_text(family = "inter", size = 12, hjust = 0, margin = margin(b = 15)), plot.caption = element_text(family = "inter", size = 8, hjust = 0, color = "gray50"), legend.position = c(1.03, 0.35), # (x, y): 1.0 = right edge, >1 = outside legend.justification = c("left", "top"), legend.box.margin = margin(0, 0, 0, 0), legend.background = element_rect(fill = "white", color = NA), legend.title = element_text(face = "bold", size = 11), legend.text = element_text(size = 10), plot.margin = margin(10, 200, 5, 10) ) map ``` ### 3.3 Plot Line Chart ```{r} line <- ggplot(km, aes(x = YEAR, y = KM)) + geom_line(color = "#2C3E50", linewidth = 0.8) + # Highlight the peak geom_vline(xintercept = 1960, linetype = "dashed", color = "#C0392B", alpha = 0.5) + scale_y_continuous( labels = scales::label_number(scale = 1/1000, suffix = "k"), breaks = seq(0, 35000, 5000) ) + scale_x_continuous(breaks = seq(1854, 2023, 30)) + labs( x = "Year", y = "Railway Extension (km)", title = NULL ) + theme_minimal() + theme( plot.background = element_rect(fill = "white", color = NA), panel.background = element_rect(fill = "white", color = NA), panel.grid.minor = element_blank(), panel.grid.major = element_line(color = "gray90", size = 0.3), axis.title = element_text(family = "inter", size = 10, face = "bold"), axis.text = element_text(family = "inter", size = 9, angle = 45), axis.title.y = element_text(angle = 90, margin = margin(r = 10)), axis.title.x = element_text(margin = margin(t = 10)), plot.margin = margin(20, 20, 20, 20) ) line ``` ```{r} # Add a light border to your line chart for clarity line_box <- line + theme( plot.background = element_rect(fill = "white", color = "gray85", linewidth = 0.3), panel.grid.major = element_line(color = "gray90", linewidth = 0.25), axis.text = element_text(size = 8, family = "inter"), axis.title = element_text(size = 9, family = "inter", face = "bold") ) final_plot <- ggdraw() + draw_plot(map, x = 0, y = 0, width = 1, height = 1) + draw_plot( line_box, x = 0.68, y = 0.35, width = 0.28, height = 0.35, hjust = 0, vjust = 0 ) final_plot ```