Creating Factors to Conflate Census Demographics to TAZ Demographics for Equity Analyses

Sophie Fox and Steven Andrews 2023-02-23

General Goal

The TDM outputs results at a TAZ level. However, for equity analyses, we need to allocated census demographics to those TAZs–the model does not output results based on minority status, income status, or any other status that might be interesting for an equity analysis (ages or limited-English proficiency, perhaps disability status, and others).

Census geometry does NOT fit neatly into TAZ geometry. Some TAZs are contained wholly within census geography, some is partially covered by multiple geographies, some is almost congruent with census geography but has slivers of overlaps.

We would like to allocate pieces of census demographics to surrounding TAZs to create demographics of the TAZs. We need to create a table that we can use to allocate portions of census geometry into TAZs.

Problems:

• TAZs are mostly made up of census blocks, but they’re not exactly made up of blocks.
• There are small tweaks to the TAZ geometry throughout the model region that mean lines do not overlap exactly.
• Census blocks are NOT necessarily the best starting point because the differential privacy features of the 2020 census have rendered them unreliable.
• TAZs may be smaller than a census geometry.

Constraints:

• TAZ geometry may change over time. The process should be repeatable for arbitrary TAZ geometry (this essentially means for ANY geometry).

• We want the process to be maintainable over multiple ACS iterations.

Other Goals

• Create multiple methods to compare the sensitivity to methodology.

Potential Solutions

When joining 2010 census blocks to TAZs, Paul Reim performed a series of intersections to identify where blocks were split. Where a TAZ split a block, he counted rooftops alongside Google StreetView to estimate how many rooftops were in each TAZ. It is not desirable to maintain such a process–the reproducibility is limited and its highly manual. We can replicate the methodology using the rooftops layer–forgoing a visual inspection methodology entirely. It is unclear exactly which dataset and how manual the process was.

Potential Solutions

1. Rooftops
   1. Allocate based on where the largest piece of the rooftop is (chosen)
   2. Where the centroid of the roof is–could be a more scalable method–assuming intersecting points and polygons is more efficient that polygons and polygons and calculating shares in each, but would want a method to keep the centroid within the geometry.
   3. Allocate proportionally based on the rooftop area-seems overly complicated
2. Land Area
   1. Find the share of the overlap removing inland water. Coastlines are clipped to match.
3. Total Area
   1. Find the share of the overlap without removing inland water. Coastlines are clipped to match.
4. Dasymetric Mapping (not completed yet)

Useful references

• massgis census: https://www.mass.gov/info-details/massgis-data-2020-us-census

• MassGIS rooftops: https://www.mass.gov/info-details/massgis-data-building-structures-2-d

• Microsoft Rooftops: https://github.com/Microsoft/USBuildingFootprints

• Dasymetric Mapping: https://www.nature.com/articles/s41597-022-01603-z

• Paul’s thoughts on the topic: TAZs and Census Geography

Running the Script

The work for this project is contained in a quarto doc named tract_to_TAZ_2020 - Final.qmd in the main directory. The document is intended to be run from top to bottom without much hand holding. Throughout the script resource-intensive operations are stored as .rds files. These statements should be turned off for a fresh run. A search for saveRDS is a way to find these places. The functions used in this document are contained in Census_to_TAZ_functions.R in the functions folder.

The user will need to download the MassGIS rooftops file and load it into the ./data/base/ folder.

The folder structure is basically a data folder with “base” untouched data, “processed” data that is a modified version of the base data or large, typically modified–datasets stored as serialized .rds files.

+-- blkgrp_to_TAZ_2020Update.html
+-- blkgrp_to_TAZ_2020Update.qmd
+-- data
|   +-- base
|   \-- processed
+-- images
+-- output
|   \-- data
+-- README.md
\-- README.qmd (generates readme.md)

Results

The final results are found in the ./output/data folder:

Area based methods: ./output/data/area_allocation_BosMPO.csv

Rooftop based method: ./output/data/rooftop_allocation_MPO_largest_in_each.csv

All methods: ./output/data/allocation_results_BosMPO.csv

Comparison of results:

The three methods tend to generate very similar results at a percentage level (there is a little more “fuzz” at the total levels). There are dozens of interesting lines that are not on the 1:1 line that we can explore in more detail.