DATA 608:Story - 7 : Where Do Strategic Minerals Come From?

drop <- c("CRIT_MIN")
piv <- c("MAJ_SOURCE_1", "MAJ_SOURCE_2", "MAJ_SOURCE_3", "MAJ_SOURCE_4")
sources_imported_as_piv <- main_df |>
    select(-all_of(drop)) |>
    distinct(IMPORTED_AS, .keep_all = TRUE) |>
    pivot_longer(cols = all_of(piv), names_to = "src_rank", values_to = "src") |>
    mutate(src_rank = as.numeric(str_replace_all(src_rank, "MAJ_SOURCE_", ""))) |>
    left_join(sources, by = join_by(src == Source),
              relationship = "many-to-one") |>
    group_by(src) |>
    mutate(src_count = n())
drop <- c("IMPORTED_AS")
sources_crit_min_piv <- main_df |>
    select(-all_of(drop)) |>
    distinct(CRIT_MIN, .keep_all = TRUE) |>
    pivot_longer(cols = all_of(piv), names_to = "src_rank", values_to = "src") |>
    mutate(src_rank = as.numeric(str_replace_all(src_rank, "MAJ_SOURCE_", ""))) |>
    left_join(sources, by = join_by(src == Source),
              relationship = "many-to-one") |>
    group_by(src) |>
    mutate(src_count = n())

drop <- c("IMPORTED_AS")
key_stats_df <- main_df |>
    select(-all_of(drop)) |>
    distinct(CRIT_MIN, .keep_all = TRUE) |>
    mutate(DEPENDENCE = case_when(DEPENDENCE == "Very High Dependence" ~ "Very High or Complete Dependence",
                                  DEPENDENCE == "Complete Dependence" ~ "Very High or Complete Dependence",
                                  TRUE ~ DEPENDENCE)) |>
    group_by(DEPENDENCE) |>
    summarize(dep_count = n()) |>
    mutate(dep_total = sum(dep_count),
           dep_percent = round(dep_count / dep_total * 100, 2)) |>
    filter(DEPENDENCE == "Very High or Complete Dependence")
cols <- c("Stat", "count", "total", "percent")
colnames(key_stats_df) <- cols
key_stats_df[] <- lapply(key_stats_df, as.character)
has_ally_source_df <- sources_crit_min_piv |>
    filter(Category == "Ally")
has_ally_source_crit_min_vec <- unique(has_ally_source_df$CRIT_MIN)
a <- c("Has Ally Source", "41", "50", as.character(round(41 / 50 * 100, 2)))
b <- c("Has No Ally Source", "6", "50", as.character(round(6 / 50 * 100, 2)))
new_rows <- as.data.frame(rbind(a, b))
rownames(new_rows) <- NULL
colnames(new_rows) <- cols
src1_df <- sources_crit_min_piv |>
    filter(src_rank == 1) |>
    group_by(src) |>
    summarize(src1_count = n()) |>
    mutate(src1_total = sum(src1_count),
           src1_percent = round(src1_count / src1_total * 100, 2))
src1_df_can_china_only <- src1_df |>
    filter(src == "Canada" | src == "China") |>
    mutate(src = ifelse(src == "Canada", "Canada Top Supplier",
                         "China Top Supplier"))
colnames(src1_df_can_china_only) <- cols
src1_df_can_china_only[] <- lapply(src1_df_can_china_only, as.character)
key_stats_df <- key_stats_df |>
    bind_rows(new_rows, src1_df_can_china_only)
sel <- c("CRIT_MIN", "IMPORTED_AS")
has_ally_source_df <- has_ally_source_df |>
    left_join(main_df |> select(sel), relationship = "many-to-one")

has_ally_source_imp_as_vec <- unique(has_ally_source_df$IMPORTED_AS)
sources_imported_as_piv <- sources_imported_as_piv |>
    mutate(Has_Ally_Src = ifelse(IMPORTED_AS %in% has_ally_source_imp_as_vec,
                                 "Yes", "No"))
palette <- brewer.pal(n = 12, name = "Paired")
fil <- c(palette[2], palette[8], greys[6])
facevec <- rev(c("bold", rep("plain", 11)))
colvec <- rev(c(palette[6], rep("black", 11)))
title_md <- "The US completely depends on imports for 12 critical minerals and <span style='color:#E31A1C;'>has no  
ally sources</span> among its top four suppliers of arsenic"
cap_str <- "Throughout, 'Ally' includes members of the Minerals Security Partnership, while 'Competitor' includes China and\nthe members of its Belt and Road Initiative. (South Korea is the only source that is a member of both, and it is\nclassified here as 'Ally.') All other sources are labeled 'Neutral.'"
library(ggrepel)

p2a <- sources_imported_as_piv |>
    filter(!is.na(src) & DEPENDENCE == "Complete Dependence") |>
    ggplot(aes(x = src_rank, y = reorder(IMPORTED_AS, desc(IMPORTED_AS)))) +
    geom_tile(aes(fill = Category), col = "white") +
    geom_text(aes(label = src), col = "white") +
    scale_fill_manual(values = fil) +
    labs(title = title_md,
         caption = cap_str,
         fill = "") +
    facet_wrap(~ DEPENDENCE, scales = "free_y", ncol = 1) + 
    theme(legend.position = "right",
          plot.title = element_markdown(),
          plot.title.position = "plot",
          plot.caption.position = "plot",
          plot.caption = element_text(hjust = 0),
          axis.title = element_blank(),
          axis.line = element_blank(),
          axis.ticks = element_blank(),
          axis.text.y = element_text(face = facevec, color = colvec))

facevec <- rev(c("bold", "bold", "plain", "bold", rep("plain", 3),
                 "bold", "plain", "plain"))
colvec <- rev(c(palette[6], palette[6], "black", palette[6], rep("black", 3),
                palette[6], "black", "black"))
title_md <- "The US depends very highly (75%+) on imports for 23 critical minerals and <span style='color:#E31A1C;'>has no  
ally sources</span> among its top four suppliers of antimony, barite, chromium, or tin"
cap_str <- "RARE EARTHS* includes aggregated import data for 14 critical minerals."
library(ggrepel)

p2b <- sources_imported_as_piv |>
  filter(!is.na(src) & DEPENDENCE == "High Dependence") |>
  ggplot(aes(x = src_rank, y = reorder(IMPORTED_AS, desc(IMPORTED_AS)))) +
  geom_tile(aes(fill = Category), color = "white") +
  geom_text_repel(aes(label = src),
                  color = "white",
                  size = 3,
                  box.padding = 0.3,
                  point.padding = 0.2,
                  max.overlaps = Inf) +
  scale_fill_manual(values = fil) +
  labs(title = title_md,
       caption = cap_str,
       fill = "") +
  facet_wrap(~ DEPENDENCE, scales = "free_y", ncol = 1) +
  theme(legend.position = "right",
        plot.title = element_markdown(),
        plot.title.position = "plot",
        plot.caption.position = "plot",
        plot.caption = element_text(hjust = 0),
        axis.title = element_blank(),
        axis.line = element_blank(),
        axis.ticks = element_blank(),
        axis.text.y = element_text(face = facevec, color = colvec))

title_md <- "The US is highly dependent (50%+) on imports for seven critical minerals,  
none of which are exclusively available from competitors or neutral parties"
p2c <- sources_imported_as_piv |>
    filter(!is.na(src) & DEPENDENCE == "High Dependence") |>
    ggplot(aes(x = src_rank, y = reorder(IMPORTED_AS, desc(IMPORTED_AS)))) +
    geom_tile(aes(fill = Category), col = "white") +
    geom_text(aes(label = src), col = "white") +
    scale_fill_manual(values = fil) +
    labs(title = title_md,
         fill = "") +
    facet_wrap(~ DEPENDENCE, scales = "free_y", ncol = 1) + 
    theme(legend.position = "right",
          plot.title = element_markdown(),
          plot.title.position = "plot",
          axis.title = element_blank(),
          axis.line = element_blank(),
          axis.ticks = element_blank())
levs <- c("Medium Dependence", "Low Dependence")
facevec <- rev(c("bold", rep("plain", 2)))
colvec <- rev(c(palette[6], rep("black", 2)))
title_md <- "The US is low (<25%) to medium (<50%) dependent on imports for four  
critical minerals and <span style='color:#E31A1C;'>has no ally sources</span> among its top four suppliers of lithium"
p2d <- sources_imported_as_piv |>
    filter(!is.na(src) & DEPENDENCE %in% levs) |>
    ggplot(aes(x = src_rank, y = reorder(IMPORTED_AS, desc(IMPORTED_AS)))) +
    geom_tile(aes(fill = Category), col = "white") +
    geom_text(aes(label = src), col = "white") +
    scale_fill_manual(values = fil) +
    labs(title = title_md,
         fill = "") +
    facet_wrap(~ factor(DEPENDENCE, levels = levs), scales = "free_y", ncol = 1) + 
    theme(legend.position = "right",
          plot.title = element_markdown(),
          plot.title.position = "plot",
          axis.title = element_blank(),
          axis.line = element_blank(),
          axis.ticks = element_blank(),
          axis.text.y = element_text(face = facevec, color = colvec))

p2b

p2c

p2d