updating to directly calculate overlay

data/data-pipeline/data_pipeline/etl/base.py

@@ -11,6 +11,7 @@ from data_pipeline.etl.score.etl_utils import (
     compare_to_list_of_expected_state_fips_codes,
 )
 from data_pipeline.etl.score.schemas.datasets import DatasetsConfig
+from data_pipeline.score import field_names
 from data_pipeline.utils import (
     load_yaml_dict_from_file,
     unzip_file_from_url,
@@ -387,6 +388,7 @@ class ExtractTransformLoad:
                 # Tract ID, but these will be ignored if they're not present.
                 cls.GEOID_FIELD_NAME: "string",
                 cls.GEOID_TRACT_FIELD_NAME: "string",
+                field_names.ZIP_CODE: "string",
             },
         )
 

data/data-pipeline/data_pipeline/etl/score/etl_score_post.py

@@ -542,7 +542,6 @@ class PostScoreETL(ExtractTransformLoad):
         # load codebook to disk
         codebook_df.to_csv(codebook_path, index=False)
 
-
         # TODO: Write zip-code based files
 
         logger.info("Compressing files")

data/data-pipeline/data_pipeline/etl/sources/geocorr_alternatives/etl.py

@@ -1,15 +1,8 @@
-import typing
-from typing import List
-
 import geopandas as gpd
-import numpy as np
 import pandas as pd
-
-from data_pipeline.config import settings
 from data_pipeline.etl.base import ExtractTransformLoad, ValidGeoLevel
 from data_pipeline.etl.sources.geo_utils import (
     add_tracts_for_geometries,
-    get_tribal_geojson,
     get_tract_geojson,
 )
 from data_pipeline.score import field_names
@@ -21,86 +14,91 @@ logger = get_module_logger(__name__)
 class GeoCorrAlternativesETL(ExtractTransformLoad):
     """Calculates overlap between Census tracts & various alternative geographies."""
 
-    # Metadata for the baseclass
     NAME = "geocorr_alternatives"
     GEO_LEVEL = ValidGeoLevel.CENSUS_TRACT
-    PUERTO_RICO_EXPECTED_IN_DATA = False
-
-    INPUT_GEOCORR_TRACT_FIELD = "tract"
-    INPUT_GEOCORR_COUNTY_FIELD = "county"
-    INPUT_GEOCORR_ZIP_FIELD = "zcta5"
-    INPUT_GEOCORR_ALLOCATION_FIELD = "afact"
+    ISLAND_AREAS_EXPECTED_IN_DATA = True
 
     # Skip some validation checks, because there will be multiple rows per tract in this
     # geocorr dataset.
     VALIDATE_SHOULD_SKIP_DUPLICATE_GEOGRAPHIES_AND_GEOGRAPHY_COUNT = True
 
-    # GeoCorr downloads have a field definition in the second row of the CSV.
-    # This parameter skips the second row for pandas `read_csv`.
-    GEOCORR_SKIP_ROWS: typing.List[int] = [1]
+    CRS_INTEGER = 3857
 
-    # GeoCorr 2010 Zip Codes (Zip Code Tabulation Area 5, ZCTA5) -> 2010 Tracts
-    # This file was generated in the UI at https://mcdc.missouri.edu/.
-    # The network request to generate it is:
-    # https://mcdc.missouri.edu/cgi-bin/broker?_PROGRAM=apps.geocorr2018.sas&_SERVICE=MCDC_long&_debug=0&state=Mo29&state=Al01&state=Ak02&state=Az04&state=Ar05&state=Ca06&state=Co08&state=Ct09&state=De10&state=Dc11&state=Fl12&state=Ga13&state=Hi15&state=Id16&state=Il17&state=In18&state=Ia19&state=Ks20&state=Ky21&state=La22&state=Me23&state=Md24&state=Ma25&state=Mi26&state=Mn27&state=Ms28&state=Mt30&state=Ne31&state=Nv32&state=Nh33&state=Nj34&state=Nm35&state=Ny36&state=Nc37&state=Nd38&state=Oh39&state=Ok40&state=Or41&state=Pa42&state=Ri44&state=Sc45&state=Sd46&state=Tn47&state=Tx48&state=Ut49&state=Vt50&state=Va51&state=Wa53&state=Wv54&state=Wi55&state=Wy56&g1_=zcta5&g2_=tract&wtvar=pop10&nozerob=1&title=&csvout=1&namoptf=b&listout=1&lstfmt=html&namoptr=b&oropt=&counties=&metros=&places=&latitude=&longitude=&locname=&distance=&kiloms=0&nrings=&r1=&r2=&r3=&r4=&r5=&r6=&r7=&r8=&r9=&r10=&lathi=&latlo=&longhi=&longlo=
-    ZIP_CODES_TO_TRACTS_PATH = (
-        settings.AWS_JUSTICE40_DATASOURCES_URL
-        + "/geocorr2018_zcta5_to_tracts.csv.zip"
+    ZCTA_2020_SHAPEFILE_PATH = (
+        "https://www2.census.gov/geo/tiger/GENZ2020/shp"
+        + "/cb_2020_us_zcta520_500k.zip"
     )
 
-    COLUMNS_TO_KEEP = [
-        ExtractTransformLoad.GEOID_TRACT_FIELD_NAME,
-        field_names.ZIP_CODE,
-        field_names.PERCENT_OF_ZIP_CODE_IN_TRACT,
-    ]
+    ZIP_CODE_INPUT_FIELD = "ZCTA5CE20"
+    AREA_JOINED_FIELD = "area_joined"
+    AREA_ZIP_FIELD = "area_zip"
 
-    # Define these for easy code completion
     def __init__(self):
-        self.ZIP_CODES_TO_TRACTS_CSV = (
-            self.get_tmp_path() / "geocorr2018_zcta5_to_tracts.csv"
-        )
+        self.COLUMNS_TO_KEEP = [
+            self.GEOID_TRACT_FIELD_NAME,
+            field_names.ZIP_CODE,
+            field_names.PERCENT_OF_ZIP_CODE_IN_TRACT,
+            self.AREA_JOINED_FIELD,
+            self.AREA_ZIP_FIELD,
+        ]
 
         self.output_df: pd.DataFrame
+        self.census_tract_gdf: gpd.GeoDataFrame
 
     def extract(self) -> None:
-        # Load the tracts to zip codes data.
+        # Download 2020 zip boundaries.
         unzip_file_from_url(
-            file_url=self.ZIP_CODES_TO_TRACTS_PATH,
+            file_url=self.ZCTA_2020_SHAPEFILE_PATH,
             download_path=self.get_tmp_path(),
-            unzipped_file_path=self.get_tmp_path(),
+            unzipped_file_path=self.get_tmp_path() / "cb_2020_us_zcta520_500k",
             verify=True,
         )
 
+        # Load census
+        self.census_tract_gdf = get_tract_geojson()
+
     def transform(self) -> None:
         logger.info("Starting GeoCorr alternatives transforms.")
-        zip_codes_to_tracts_df = pd.read_csv(
-            filepath_or_buffer=self.ZIP_CODES_TO_TRACTS_CSV,
-            dtype={
-                self.INPUT_GEOCORR_TRACT_FIELD: "string",
-                self.INPUT_GEOCORR_COUNTY_FIELD: "string",
-                self.INPUT_GEOCORR_ZIP_FIELD: "string",
-            },
-            skiprows=self.GEOCORR_SKIP_ROWS,
-        )
 
-        zip_codes_to_tracts_df = zip_codes_to_tracts_df.rename(
-            columns={
-                self.INPUT_GEOCORR_ALLOCATION_FIELD: field_names.PERCENT_OF_ZIP_CODE_IN_TRACT,
-                self.INPUT_GEOCORR_ZIP_FIELD: field_names.ZIP_CODE,
-            },
+        # Read in ZCTA data.
+        zcta_2020_gdf = gpd.read_file(
+            filename=self.get_tmp_path() / "cb_2020_us_zcta520_500k"
+        )
+        zcta_2020_gdf = zcta_2020_gdf.rename(
+            columns={self.ZIP_CODE_INPUT_FIELD: field_names.ZIP_CODE},
             errors="raise",
         )
 
-        # Create the tract ID by combining fields from GeoCorr.
-        zip_codes_to_tracts_df[self.GEOID_TRACT_FIELD_NAME] = (
-            zip_codes_to_tracts_df[self.INPUT_GEOCORR_COUNTY_FIELD]
-            + zip_codes_to_tracts_df[self.INPUT_GEOCORR_TRACT_FIELD]
+        # Switch from geographic to projected CRSes
+        # because logically that's right
+        self.census_tract_gdf = self.census_tract_gdf.to_crs(
+            crs=self.CRS_INTEGER
         )
-        # Remove unnecessary periods.
-        zip_codes_to_tracts_df[
-            self.GEOID_TRACT_FIELD_NAME
-        ] = zip_codes_to_tracts_df[self.GEOID_TRACT_FIELD_NAME].str.replace(
-            ".", "", regex=False
+        zcta_2020_gdf = zcta_2020_gdf.to_crs(crs=self.CRS_INTEGER)
+
+        # Calculate percentage overlap.
+        # Create a measure of area for the entire Zip area.
+        zcta_2020_gdf[self.AREA_ZIP_FIELD] = zcta_2020_gdf.area
+
+        # Perform overlay function.
+        # We have a mix of polygons and multipolygons, and we just want the overlaps
+        # without caring a ton about the specific types, so we ignore geom type.
+        # Realistically, this changes almost nothing in the calculation; True and False
+        # are the same within 9 digits of precision.
+        joined_gdf = gpd.overlay(
+            df1=zcta_2020_gdf,
+            df2=self.census_tract_gdf,
+            how="intersection",
+            keep_geom_type=False,
         )
 
-        self.output_df = zip_codes_to_tracts_df
+        # Calculating the areas of the newly-created overlapping geometries
+        joined_gdf[self.AREA_JOINED_FIELD] = joined_gdf.area
+
+        # Calculating the areas of the newly-created geometries in relation
+        # to the original tract geometries
+        joined_gdf[field_names.PERCENT_OF_ZIP_CODE_IN_TRACT] = (
+            joined_gdf[self.AREA_JOINED_FIELD] / joined_gdf[self.AREA_ZIP_FIELD]
+        )
+
+        self.output_df = joined_gdf