Open Climate Risk (OCR) Beta

This Source Cooperative repository contains data products for the beta release of the carbonplan-ocr project. These data were used to build a scalable pipeline for calculating climate risk assessments at building-level resolution across the Continental United States. Wildfire risk was calculated through a distributed processing system that can run locally or on cloud infrastructure. Technical documentation can be found here.

Working with the data

Documentation for working with the raster and vector data can be found here.

Data Formats

Tensor

• Our raw input data consisted of either Zarr or TIFF files. We processed it into rechunked Icechunk stores.
• Our n-dimensional output data is stored as a single Icechunk store.

Vector

• Our raw input data was either GPKG or Geoparquet. All the vector pipeline inputs were converted to Geoparquet.
• Our output vector data is stored in multiple formats. See our documentation on data downloads for more details.
  • Hive-partitioned (state/county) Geoparquet file for risk sampled buildings across CONUS.
  • PMTiles layers for buildings, Census blocks, Census Tracts and counties.
  • Summary statistics files in GPKG and CSV

Data structure

The data is split into inputs and outputs, where inputs are raw and processed data for the pipeline, and outputs are versioned dataset outputs.

Terms of use

By viewing Open Climate Risk data, you agree to CarbonPlan’s Terms of Use and Privacy Policy, and that the data is in a beta state and may not be used for decision-making purposes, cited, or otherwise shared.

This work contains information from the Overture Maps buildings database, which is made available here under the Open Database License (ODbL).

Licenses for tensor (CC BY 4.0) and vector (ODbL) data are packaged with the data in colocated LICENSE.txt files.