Downloaded the City of Austin jurisdictions shapefile here: https://data.austintexas.gov/City-Government/BOUNDARIES_jurisdictions/3pzb-6mbr
Downloaded Texas census tracts shapefile from TIGER/line here: https://www.census.gov/geographies/mapping-files/time-series/geo/tiger-line-file.html
I am filtering for only full and limited purpose jurisdictions.
I decided to do this analysis in R because performing an intersecting overlay or spatial join in ArcGIS would have cost 84 credits.
library(sf)Linking to GEOS 3.13.1, GDAL 3.11.0, PROJ 9.6.0; sf_use_s2() is TRUElibrary(readr)
library(dplyr)
Attaching package: 'dplyr'The following objects are masked from 'package:stats':
filter, lagThe following objects are masked from 'package:base':
intersect, setdiff, setequal, union# read Austin jurisdictions shapefile
austin_jur <- st_read("BOUNDARIES_jurisdictions_20260205/geo_export_5db1be39-2392-451a-8502-b3021a29112a.shp")Reading layer `geo_export_5db1be39-2392-451a-8502-b3021a29112a' from data source `C:\Users\kaitl\OneDrive\Documents\Every Texan\R\City of Austin Equity Index\Full and LTD census tracts\BOUNDARIES_jurisdictions_20260205\geo_export_5db1be39-2392-451a-8502-b3021a29112a.shp'
using driver `ESRI Shapefile'
Simple feature collection with 352 features and 10 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: -98.01504 ymin: 30.04298 xmax: -97.48034 ymax: 30.51685
Geodetic CRS: WGS 84# quick check
print(unique(austin_jur$JURISDICTION_LABEL))NULL# inspect Austin shapefile
names(austin_jur) [1] "objectid" "jurisdicti" "city_name" "jurisdic_2" "modified_f"
[6] "jurisdic_3" "jurisdic_4" "shape_area" "shape_leng" "globalid"
[11] "geometry" head(austin_jur)Simple feature collection with 6 features and 10 fields
Geometry type: MULTIPOLYGON
Dimension: XY
Bounding box: xmin: -97.92222 ymin: 30.22973 xmax: -97.77758 ymax: 30.3428
Geodetic CRS: WGS 84
objectid jurisdicti city_name jurisdic_2 modified_f
1 312 401472386 CITY OF AUSTIN AUSTIN 2 MILE ETJ 202
2 230 401472355 CITY OF AUSTIN AUSTIN 2 MILE ETJ 202
3 231 401472356 CITY OF AUSTIN AUSTIN 2 MILE ETJ 202
4 258 401472326 CITY OF AUSTIN AUSTIN 2 MILE ETJ 202
5 193 10 CITY OF AUSTIN AUSTIN LTD 211
6 222 401472320 CITY OF AUSTIN AUSTIN 2 MILE ETJ 202
jurisdic_3 jurisdic_4 shape_area shape_leng
1 DISANNEXED PER SB 1844 2 MILE ETJ 7263.191 467.3437
2 DISANNEXED PER SB 1844 2 MILE ETJ 43172.439 1007.8005
3 DISANNEXED PER SB 1844 2 MILE ETJ 61576.410 1022.9614
4 DISANNEXED PER SB 1844 2 MILE ETJ 43322.393 837.6653
5 LIMITED PURPOSE PLANNING ZONING LTD 7173169.650 16879.2008
6 DISANNEXED PER SB 1844 2 MILE ETJ 67123.084 1494.6134
globalid geometry
1 2644acde-7560-4f95-9336-d04480bb6ba6 MULTIPOLYGON (((-97.88191 3...
2 b5fef16f-58ac-4f6d-8757-bc27ce043d4c MULTIPOLYGON (((-97.78378 3...
3 eb5ea736-435c-49d0-8318-5c2cb4f49fb7 MULTIPOLYGON (((-97.8187 30...
4 150cc82f-8b67-4924-a6e8-0d0f07d97ddd MULTIPOLYGON (((-97.92192 3...
5 dabb5d59-05a0-4ea2-96a0-a69303af5c83 MULTIPOLYGON (((-97.899 30....
6 4fac79bf-020e-4d9b-b655-dc8574a7852b MULTIPOLYGON (((-97.77912 3...Here I am filtering to include only limited and full-purpose jurisdictions in Austin.
austin_jur_filtered <- austin_jur %>%
filter(jurisdic_2 %in% c("AUSTIN LTD", "AUSTIN FULL PURPOSE"))
print(unique(austin_jur_filtered$jurisdic_2))[1] "AUSTIN LTD" "AUSTIN FULL PURPOSE"# read the Texas Census tracts shapefile
tx_census_tracts <- st_read("tl_2025_48_tract/tl_2025_48_tract.shp")Reading layer `tl_2025_48_tract' from data source
`C:\Users\kaitl\OneDrive\Documents\Every Texan\R\City of Austin Equity Index\Full and LTD census tracts\tl_2025_48_tract\tl_2025_48_tract.shp'
using driver `ESRI Shapefile'
Simple feature collection with 6896 features and 13 fields
Geometry type: POLYGON
Dimension: XY
Bounding box: xmin: -106.6456 ymin: 25.83705 xmax: -93.50804 ymax: 36.5007
Geodetic CRS: NAD83# check columns
print(names(tx_census_tracts)) [1] "STATEFP" "COUNTYFP" "TRACTCE" "GEOID" "GEOIDFQ" "NAME"
[7] "NAMELSAD" "MTFCC" "FUNCSTAT" "ALAND" "AWATER" "INTPTLAT"
[13] "INTPTLON" "geometry"# rebuild a clean GEOID (I was getting mostly NA values for some reason for GEOID)
tx_census_tracts <- tx_census_tracts %>%
mutate(
GEOID_clean = paste0(STATEFP, COUNTYFP, TRACTCE)
)
# check
head(tx_census_tracts$GEOID_clean)[1] "48201342001" "48201421101" "48201423301" "48381020400" "48029171926"
[6] "48113017819"# double check
#unique(census_tracts$GEOID_clean)
# inspect new GEOIDs in Excel
tract_ids <- unique(tx_census_tracts$GEOID_clean)
write.csv(tract_ids, "All_Census_Tract_GEOIDs.csv", row.names = FALSE)
# there are 6,896 tracts in Texas according to this list# make sure both layers have valid geometry
austin_jur_filtered <- st_make_valid(austin_jur_filtered)
tx_census_tracts <- st_make_valid(tx_census_tracts)# match coordinate reference systems
if (st_crs(tx_census_tracts) != st_crs(austin_jur_filtered)) {
tx_census_tracts <- st_transform(tx_census_tracts, st_crs(austin_jur_filtered))
}# spatially join tracts to any Austin jurisdiction polygon they intersect
selected_tracts <- st_join(tx_census_tracts, austin_jur_filtered, join = st_intersects)# keep only those that actually intersect Austin (non-NA jurisdiction label)
selected_tracts <- selected_tracts %>%
filter(!is.na(jurisdic_2))# check for NA values
table(is.na(selected_tracts$GEOID_clean))
FALSE
383 head(selected_tracts$GEOID_clean, 20) [1] "48453002112" "48453002113" "48453002201" "48453041300" "48453031600"
[6] "48453032500" "48453045600" "48453045600" "48453043000" "48453043000"
[11] "48453043000" "48453034200" "48453042000" "48453042000" "48453042000"
[16] "48453045100" "48453045100" "48453041400" "48453032100" "48453043100"# quick check that the selected tracts look reasonable
print(paste("# of tracts intersecting Austin:", nrow(selected_tracts)))[1] "# of tracts intersecting Austin: 383"print(paste("# of unique GEOIDs:", length(unique(selected_tracts$GEOID_clean))))[1] "# of unique GEOIDs: 271"There are 271 intersecting census tracts, meaning there are 271 tracts in Austin full and limited purpose jurisdictions. This makes sense, as Travis County alone has 291 census tracts, though not all of Travis County is in Austin. There are a total of 493 census tracts in all of Bastrop, Hays, Travis, and Williamson Counties combined.
The reason there are 383 rows, but only 271 distinct tracts, could be because some tracts may touch multiple Austin jurisdiction polygons since Austin’s boundary is split into parts.
# quick view of unique GEOIDs
print(sort(unique(selected_tracts$GEOID_clean))[1:20]) [1] "48021950303" "48209010912" "48209010923" "48453000101" "48453000102"
[6] "48453000203" "48453000204" "48453000205" "48453000206" "48453000302"
[11] "48453000304" "48453000305" "48453000307" "48453000308" "48453000309"
[16] "48453000401" "48453000402" "48453000500" "48453000601" "48453000605"# to save only distinct GEOIDs with no duplicates
selected_unique <- selected_tracts %>%
distinct(GEOID_clean, .keep_all = TRUE)# save the intersected tracts shapefile
st_write(selected_tracts, "Austin_LTD_FULL_Tracts.shp")Warning in abbreviate_shapefile_names(obj): Field names abbreviated for ESRI
Shapefile driverWriting layer `Austin_LTD_FULL_Tracts' to data source
`Austin_LTD_FULL_Tracts.shp' using driver `ESRI Shapefile'
Writing 383 features with 24 fields and geometry type Polygon.# Also export a CSV list of tract GEOIDs and names
tract_list <- selected_tracts %>%
st_drop_geometry() %>%
arrange(GEOID_clean)
write_csv(tract_list, "Austin_LTD_FULL_Tract_List.csv")
###################################################################################
# files without duplicate GEOIDs
st_write(selected_unique, "Austin_LTD_FULL_Tracts_Unique.shp")Warning in abbreviate_shapefile_names(obj): Field names abbreviated for ESRI
Shapefile driverWriting layer `Austin_LTD_FULL_Tracts_Unique' to data source
`Austin_LTD_FULL_Tracts_Unique.shp' using driver `ESRI Shapefile'
Writing 271 features with 24 fields and geometry type Polygon.unique_tract_list <- selected_unique %>%
st_drop_geometry() %>%
arrange(GEOID_clean)
write_csv(unique_tract_list, "Austin_LTD_FULL_Tract_List_Unique.csv")Checking to see if I still get 271 tracts when I join my shapefile to Coda’s shapefile. Coda’s shapefile contains all census tracts in Bastrop, Hays, Travis, and Williamson counties.
# read CRG’s shapefile
CRG_tracts <- st_read("CRG_Census_Tracts_Final_All_Counties_Shape/MergedFeatures.shp")Reading layer `MergedFeatures' from data source
`C:\Users\kaitl\OneDrive\Documents\Every Texan\R\City of Austin Equity Index\Full and LTD census tracts\CRG_Census_Tracts_Final_All_Counties_Shape\MergedFeatures.shp'
using driver `ESRI Shapefile'
Simple feature collection with 492 features and 13 fields
Geometry type: POLYGON
Dimension: XY
Bounding box: xmin: -10942440 ymin: 3471769 xmax: -10800710 ymax: 3620342
Projected CRS: WGS 84 / Pseudo-Mercatornames(CRG_tracts) [1] "STATE_ABBR" "STATE_FIPS" "COUNTY_FIP" "STCOFIPS" "TRACT_FIPS"
[6] "FIPS" "POPULATION" "POP_SQMI" "SQMI" "POPULATI_1"
[11] "POP20_SQMI" "Shape__Are" "Shape__Len" "geometry" # reconstruct clean GEOID
CRG_tracts <- CRG_tracts %>%
mutate(GEOID_clean = paste0(STATE_FIPS, COUNTY_FIP, TRACT_FIPS))
head(CRG_tracts$GEOID_clean)[1] "48453001910" "48453002105" "48453035900" "48453000305" "48453000608"
[6] "48453033300"# check which of my 271 unique GEOIDs are in Coda's file
common_geo <- intersect(unique(selected_unique$GEOID_clean),
unique(CRG_tracts$GEOID_clean))
setdiff(unique(selected_unique$GEOID_clean), unique(CRG_tracts$GEOID_clean)) # not in CRG's filecharacter(0)length(CRG_tracts$GEOID_clean)[1] 492# how many unique GEOID's in Coda's file
length(unique(CRG_tracts$GEOID_clean))[1] 492length(common_geo)[1] 271Verified that all 271 tracts are in Coda’s file.