# UCRI California CA_2026.10: reproducibility package This package lets anyone check the California Urban Canopy Risk Index (UCRI) release CA_2026.10 without Canopulse software. It contains every California 2020 census tract with its five component scores, its released score, class and rank, and the raw tract values behind them, plus three scripts: a short check that recomputes the released scores from the components, the scoring logic, and the weighting-sensitivity checks. Prepared by Canopulse. This is not an independent review; it is material an independent reviewer can use to check the released scores. The raster pipeline that summarizes canopy, impervious surface and land-surface temperature by tract is not published. ## Files | File | What it is | |---|---| | `ucri_ca_2026_10_tracts.csv` | 9,129 tracts: components, UCRI, provisional estimate, class, statewide rank, raw inputs | | `METHOD.md` | Exact method specification and how to reproduce | | `reproduce.py` | Python 3 script, standard library only, that recomputes the release from the CSV | | `compute_ucri.py` | Python 3 script, standard library only, with the scoring logic: component percentiles, UCRI, provisional estimates, classes and ranks from tract input tables (the raster-derived tables are not in this package) | | `robustness/ucri_robustness.py` | Python 3 script, standard library only, that runs the weighting-sensitivity checks and writes the three files below | | `robustness/REPORT.md` | Canopulse's internal weighting-sensitivity checks, in plain English | | `robustness/results.json` | Machine-readable results of those checks | | `robustness/tract_stability.csv` | Per tract: share of 1,000 random weightings that place it in the top 20% | | `LICENSE.md` | License: CC BY 4.0 for the data and text, MIT for the three scripts | | `CITATION.cff` | Machine-readable citation for reference managers and code repositories | | `SHA256SUMS.txt` | SHA-256 checksums for every file above | Quick start: `python3 reproduce.py` (Python 3.8 or later, no packages). The expected output is recorded in section 12 of `METHOD.md`. To rerun the weighting checks: `python3 robustness/ucri_robustness.py --out rerun` (about a minute; writes fresh copies of the robustness files to `rerun/`). ## Sources and license The tract values are derived from public government data: | Layer | Source | License | |---|---|---| | Tract, county and place geography | [US Census TIGER/Line 2020](https://www.census.gov/geographies/mapping-files/time-series/geo/tiger-line-file.html) | Public domain (US Census) | | Tree canopy | [NLCD Tree Canopy Cover 2025, v2025-6](https://data.fs.usda.gov/geodata/rastergateway/treecanopycover/), USDA Forest Service (distributed through MRLC) | Public domain | | Impervious surface | [Annual NLCD Collection 1.2 Fractional Impervious Surface 2025](https://www.usgs.gov/centers/eros/science/usgs-eros-archive-land-cover-annual-nlcd-collection-12-fractional-impervious) | Public domain (USGS) | | Land-surface temperature | [Landsat 8/9 Collection 2 Level-2](https://www.usgs.gov/landsat-missions/landsat-data-access), accessed through [Microsoft Planetary Computer](https://planetarycomputer.microsoft.com/dataset/landsat-c2-l2) | Public domain (USGS) | | Social vulnerability | [CDC/ATSDR SVI 2022](https://www.atsdr.cdc.gov/place-health/php/svi/svi-data-documentation-download.html) | Public domain (US government) | | Environmental burden | [CalEnviroScreen 5.0](https://oehha.ca.gov/calenviroscreen/maps-data), Pollution Burden percentile | No license specified on the data.ca.gov dataset page (checked October 11, 2026); published by OEHHA | Canopulse's outputs in this package (the tract scores, classes, ranks and derived values in the CSV, `METHOD.md`, this README and the report and result files in `robustness/`) are licensed under the [Creative Commons Attribution 4.0 International license (CC BY 4.0)](https://creativecommons.org/licenses/by/4.0/). You may share and adapt them for any purpose, including commercially, if you credit Canopulse, link the license and indicate any changes. The three scripts, `reproduce.py`, `compute_ucri.py` and `robustness/ucri_robustness.py`, are licensed under the MIT License. `LICENSE.md` has both. The source data keep their own licenses, listed above. Please credit the original data providers as well when you reuse the values. The Canopulse name and logo are not licensed. Keep this notice with the material: Screening context only. Not a legal, regulatory, engineering, funding or eligibility determination. Provided as is, without warranty. The UCRI scores are a screening index, not a measure of health risk, eligibility or project outcomes. ## Citation Please cite the dataset as: > Canopulse (2026). California UCRI: Urban Canopy Risk Index for California census tracts, release CA_2026.10 [Data set]. https://www.canopulse.com. CC BY 4.0. BibTeX: ```bibtex @misc{canopulse_ucri_ca_2026_10, author = {{Canopulse}}, title = {California UCRI: Urban Canopy Risk Index for California census tracts, release CA_2026.10}, year = {2026}, howpublished = {\url{https://www.canopulse.com}}, note = {Data set. Licensed under CC BY 4.0}, } ``` `CITATION.cff` carries the same citation in Citation File Format. Credit lines for maps, tables and GIS layers are at https://www.canopulse.com/cite.