feat: Add neighborhoods_2020 + baltimore_census_xwalk data

NEWS.md

@@ -2,7 +2,8 @@
 
 ## New data
 
-* Added `buildings_21stc` (2023-04-28), `main_streets` (2023-10-16), and `rec_centers` (2023-10-19)
+* Added `buildings_21stc` (2023-04-28), `main_streets` (2023-10-16), `rec_centers` (2023-10-19), and `neighborhoods_2020` (2023-11-06) spatial data.
+* Added `baltimore_area_xwalk` (2023-11-06) reference data.
 
 ## Updated data
 
@@ -15,7 +16,7 @@
 
 * Add `get_neighborhood()` and refactor `get_baltimore_area()` for more consistency with `{getdata}` parameter names (2023-05-26).
 
-## other
+## Other
 
 * Update package logo and switch pkgdown site to rendering with GitHub actions (2023-06-13)
 

R/data.R

@@ -50,7 +50,7 @@
 #' by local community or neighborhood associations as an area of responsibility
 #' or membership recruitment.
 #'
-#' @format A data frame with 278 rows and 2 variables:
+#' @format A data frame with 278 rows and 6 variables:
 #' \describe{
 #'   \item{`name`}{Neighborhood name}
 #'   \item{`type`}{Type of area, with options including residential, industrial area, park/open space, institutionl area and business park)}
@@ -62,6 +62,28 @@
 #' @source [Maryland Baltimore City Neighborhoods (MD iMap)](https://data.imap.maryland.gov/datasets/fc5d183b20a145009eae8f8b171eeb0d_0)
 "neighborhoods"
 
+
+#' Neighborhood Boundaries for Baltimore City (2020)
+#'
+#' Baltimore City neighborhoods (officially known as Neighborhood Statistical
+#' Areas) established by the Baltimore City Department of Planning based on the
+#' 2020 U.S. Decennial Census. This is an updated version of the 2010
+#' Neighborhood Statistical Areas.
+#'
+#' @format A data frame with 279 rows and 8 variables:
+#' \describe{
+#'   \item{`name`}{Neighborhood name}
+#'   \item{`name_alt`}{2010 neighborhood name}
+#'   \item{`type`}{Type of area, with options including residential, industrial area, park/open space, institutionl area and business park)}
+#'   \item{`acres`}{Area of the neighborhood (acres)}
+#'   \item{`osm_id`}{Open Street Map (OSM) relation identifier}
+#'   \item{`wikidata`}{Wikidata entity identifier}
+#'   \item{`color_id`}{Color identifier}
+#'   \item{`geometry`}{MULITPOLYGON boundary geometry}
+#' }
+#' @source [NSA_Feb2023_service](https://services1.arcgis.com/43Lm3JYE3nM91DAF/arcgis/rest/services/NSA_March2023/FeatureServer/0)
+"neighborhoods_2020"
+
 #' U.S. Census Block-to-Tract Crosswalk with 2010 Block Household Population
 #'
 #' A crosswalk file used to generate `xwalk_neighborhood2tract`.
@@ -1169,3 +1191,24 @@
 #'}
 #' @source <https://services1.arcgis.com/UWYHeuuJISiGmgXx/arcgis/rest/services/recreationCenter2023/FeatureServer>
 "rec_centers"
+
+#' Crosswalk for Baltimore areas and Census geography
+#'
+#' A pre-built crosswalk data frame that can be filtered by geography and then
+#' used with the [getACS::use_area_xwalk()] function. Crosswalk uses the 2010
+#' City Council district boundaries, 2010 and 2020 neighborhood (neighborhood
+#' statistical area) boundaries, and current city planning districts.
+#'
+#' @format A data frame with 1024 rows and 8 variables:
+#' \describe{
+#'   \item{`geography`}{Geography/area type}
+#'   \item{`GEOID`}{2020 Census GeoID}
+#'   \item{`TRACTCE20`}{2020 Census Tract ID}
+#'   \item{`name`}{Area name}
+#'   \item{`POP20`}{Population in area and tract}
+#'   \item{`perc_POP20`}{Percent of population in area and tract}
+#'   \item{`HOUSING20`}{Households/occupied housing units in area and tract}
+#'   \item{`perc_HOUSING20`}{Percent of households in area and tract}
+#'}
+#' @details Created using the [getACS::make_area_xwalk()] data.
+"baltimore_census_xwalk"

data-raw/baltimore_census_xwalk.R

@@ -0,0 +1,74 @@
+## code to prepare `area_xwalk` dataset goes here
+
+library(getACS)
+library(tidyverse)
+
+block_xwalk <- make_block_xwalk(
+  "MD",
+  "Baltimore city"
+  )
+
+area_xwalk_list_housing <- map(
+  list(
+    mapbaltimore::council_districts,
+    mapbaltimore::neighborhoods,
+    neighborhoods_2020,
+    mapbaltimore::planning_districts
+  ),
+  \(x) {
+    make_area_xwalk(
+      x,
+      block_xwalk,
+      coverage = FALSE,
+      name_col = "name",
+      weight_col = "HOUSING20"
+    )
+  }
+)
+
+area_xwalk_housing <- purrr::list_rbind(
+  set_names(
+    area_xwalk_list_housing,
+    c("council district 2010",
+      "neighborhood 2010",
+      "neighborhood 2020",
+      "planning district")
+    ),
+  names_to = "geography"
+)
+
+area_xwalk_list_pop <- map(
+  list(
+    mapbaltimore::council_districts,
+    mapbaltimore::neighborhoods,
+    neighborhoods_2020,
+    mapbaltimore::planning_districts
+  ),
+  \(x) {
+    make_area_xwalk(
+      x,
+      block_xwalk,
+      coverage = FALSE,
+      name_col = "name",
+      weight_col = "POP20"
+    )
+  }
+)
+
+area_xwalk_pop <- purrr::list_rbind(
+  set_names(
+    area_xwalk_list_pop,
+    c("council district 2010",
+      "neighborhood 2010",
+      "neighborhood 2020",
+      "planning district")
+  ),
+  names_to = "geography"
+)
+
+baltimore_census_xwalk <- dplyr::full_join(
+  area_xwalk_pop,
+  area_xwalk_housing
+)
+
+usethis::use_data(baltimore_census_xwalk, overwrite = TRUE)

data-raw/data.R

@@ -178,6 +178,83 @@ planning_districts <- planning_districts %>%
 
 usethis::use_data(planning_districts, overwrite = TRUE)
 
+neighborhoods_2020_path <- "https://services1.arcgis.com/43Lm3JYE3nM91DAF/arcgis/rest/services/NSA_March2023/FeatureServer/0"
+
+neighborhoods_2020_source <- neighborhoods_2020_path |>
+  esri2sf::esri2sf(crs = selected_crs) %>%
+  sfext::rename_sf_col()
+
+neighborhoods_2020 <- neighborhoods_2020_source |>
+  janitor::clean_names("snake") %>%
+  dplyr::mutate(
+    acres = as.numeric(units::set_units(sf::st_area(geometry), "acres")),
+    type = dplyr::case_when(
+      (stringr::str_detect(name, "Industrial") | name == "Jones Falls Area" | name == "Dundalk Marine Terminal") ~ "Industrial area",
+      stringr::str_detect(name, "Business Park") ~ "Business park",
+      name %in% c("University Of Maryland", "Morgan State University") ~ "Institutional area",
+      # NOTE: This classifies Montebello as a park but is more accurately described as a reservoir
+      name %in% c(
+        "Gwynns Falls/Leakin Park", "Druid Hill Park", "Patterson Park", "Clifton Park", "Carroll Park",
+        "Montebello", "Greenmount Cemetery", "Herring Run Park", "Lower Herring Run Park"
+      ) ~ "Park/open space",
+      TRUE ~ "Residential"
+    )
+  ) %>%
+  dplyr::select(
+    name,
+    type,
+    acres,
+    color_id = color_2,
+    geometry
+  ) %>%
+  dplyr::mutate(
+    name_alt = case_match(
+      name,
+      "Perkins" ~ "Perkins Homes",
+      "Mount Clare" ~ "New Southwest/Mount Clare",
+      "Stadium/Entertainment Area" ~ "Stadium Area",
+      "Baltimore Peninsula" ~ "Port Covington",
+      "Fairmount" ~ "Fairmont",
+      "Butchers Hill" ~ "Butcher's Hill",
+      "Villages of Homeland" ~ "Villages Of Homeland",
+      "Auchentoroly-Parkwood" ~ "Parkview/Woodbrook",
+      "Parkway" ~ "Burleith-Leighton",
+      # "Chinquapin Park" ~ "Wrenlane",
+      "Concerned Citizens of Forest Park" ~ "Concerned Citizens Of Forest Park",
+      "Carroll - Camden Industrial Area" ~ name,
+      "University Of Maryland" ~ name,
+      "Four By Four" ~ name,
+      .default = NA_character_
+      ),
+    name_join = coalesce(name_alt, name),
+    .after = name
+  ) |>
+    dplyr::mutate(
+      name = case_match(name,
+                        "University Of Maryland" ~ "University of Maryland",
+                        "Four By Four" ~ "Four by Four",
+                        "Carroll - Camden Industrial Area" ~ "Carroll-Camden Industrial Area",
+                        .default = name)
+    ) |>
+  dplyr::left_join(
+    mapbaltimore::neighborhoods |>
+      sf::st_drop_geometry() |>
+      dplyr::select(
+        osm_id,
+        wikidata,
+        name_join = name
+      ),
+    by = "name_join"
+  ) |>
+  dplyr::relocate(
+    color_id,
+    .after = "wikidata"
+  ) |>
+  dplyr::select(!name_join) |>
+  dplyr::arrange(name)
+
+usethis::use_data(neighborhoods_2020, overwrite = TRUE)
+
 # Import neighborhood boundaries from iMAP (derived from a data set previously available on Open Baltimore)
 # https://opendata.maryland.gov/Society/MD-iMAP-Maryland-Baltimore-City-Neighborhoods/dbbp-8u4u
 neighborhoods_2010_path <- "https://opendata.arcgis.com/datasets/fc5d183b20a145009eae8f8b171eeb0d_0.geojson"