# Climate Change Risk Index for 1,455 US Municipal Drinking Water Utilities (CMU KiltHub)

## Facts

- Provider: Zia Lyle, Constantine Samaras, Jeanne VanBriesen (Carnegie Mellon University)
- License: CC-BY-4.0 (https://api.figshare.com/v2/articles/28169504)
- Size: 4 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/CMU-OSPO/Climate_Change_Risk_Index_Values_for_Medium-_and_Large-_Municipal_Drinking_Water_Utilities
- Access: public, readable without an account
- DOI: https://doi.org/10.1184/R1/28169504.v2
- Category: Climate & Weather
- Source: https://huggingface.co/datasets/CMU-OSPO/Climate_Change_Risk_Index_Values_for_Medium-_and_Large-_Municipal_Drinking_Water_Utilities
- Page: https://scigantic.com/archive/a4c894cc-f072-5237-a809-f26ad8b2ae53
- Keywords: Climate & Weather, Parquet, Hugging Face

## About

Climate hazard projections, exposure and vulnerability indicators, and a comparative climate risk index for 1,455 medium and large municipal drinking water systems in the contiguous United States, compiled at Carnegie Mellon University from public sources. Each row is one public water system, keyed by its EPA Public Water System ID (pwsid), with population served, service connections, primary source water type and NOAA climate region. The 49 variables include historical and mid-century RCP4.5 projections of 5-day maximum precipitation, 5-day maximum temperature, average temperature and precipitation, freeze-thaw cycles, wildfire risk, a sea level rise indicator, energy demand and water stress, plus financial and social vulnerability measures such as average bond rating, operating ratio, water quality noncompliance score, median housing age, poverty rate and annual service cost. These feed hazard, exposure, vulnerability and overall risk indices. The Hugging Face copy, made by CMU's Open Source Programs Office, is one Parquet table that loads with the datasets library or pandas; the authors' README with variable definitions is on KiltHub. License CC BY 4.0. Cite Lyle, Samaras and VanBriesen (2025), Carnegie Mellon University, DOI 10.1184/R1/28169504.v2.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("CMU-OSPO/Climate_Change_Risk_Index_Values_for_Medium-_and_Large-_Municipal_Drinking_Water_Utilities")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `dwutils_climateriskindex_dataset/train-00000-of-00001.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/a4c894cc-f072-5237-a809-f26ad8b2ae53
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/a4c894cc-f072-5237-a809-f26ad8b2ae53
- Over MCP: `get_schema_card` with id `a4c894cc-f072-5237-a809-f26ad8b2ae53` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Climate & Weather datasets](https://scigantic.com/topics/climate-and-weather)
