2026 state RPS / CES / hydrofrac data update

[Yunzhi-Chen]

Summary

Updates the three state-policy input CSVs in inputs/state_policies/ to the 2026 vintage produced by ReEDS_Input_Processing#11.

Technical details

Three files updated in-place (same schema, refreshed values):

• inputs/state_policies/rps_fraction.csv
• inputs/state_policies/ces_fraction.csv
• inputs/state_policies/hydrofrac_policy.csv

All three CSVs are direct outputs of state_policies/data_processing.py in the input-processing repo. See ReEDS_Input_Processing#11 for the full methodology, source-data changes, and old-vs-new comparison plots.

Implementation notes

Upstream data sources refreshed in the input-processing repo:

1. LBNL state RPS / CES dataset (Galen Barbose): RPS & CES Targets and Demand_June 2025.xlsx → RPS & CES Targets and Demand_June 2026.xlsx.
2. NLR Green Power Data (voluntary RPS row): nrel-green-power-data-v2023.xlsx → nrel-green-power-data-v2024.xlsx.

Notable state-level RPS changes flow through from the LBNL June 2026 release for AZ, CT, ME, NC, NY, and VA. The voluntary trajectory shifts uniformly downward due to NLR's revised market-wide estimates in the 2024 release.

Issues resolved

Closes #122

Relevant sources or documentation

• LBNL Renewables Portfolio Standards project (annual dataset and summary slides): https://emp.lbl.gov/projects/renewables-portfolio
• LBNL June 2026 summary slides (explains the AZ, CT, ME, NC, NY, VA changes): https://emp.lbl.gov/sites/default/files/2026-06/State%20Electricity%20Resource%20Standards-2026%20Data%20Update.pdf
• NLR Green Power Data (formerly NREL): https://www.nlr.gov/analysis/voluntary-power-procurement

Validation, testing, and comparison report(s)

Update 2026 state policy (PR0) vs current 2025 version (main0)

Details to double-check

• Charge code provided to reviewers
• Included comparison reports for appropriate test cases (full-US reference and full-US decarb)
• Documentation updated if necessary — no model docs affected; input-folder README is updated in ReEDS_Input_Processing#11
• If input data added/modified:
  • Dollar year recorded and converted to 2004$ for GAMS — N/A, fractions only
  • Timeseries are in Central Time — N/A, annual values
  • Units are specified — dimensionless fractions, same as prior file
  • Preprocessing steps have been documented and committed to ReEDS_Input_Processing — see PR Standardize the definition of seasons #11
  • New large data files handled with .h5 instead of .csv — N/A, existing small CSVs updated in place
  • If spatially resolved inputs are modified, before/after/diff maps — N/A (state-level scalars; see comparison plots in ReEDS_Input_Processing#11)
• Code formatting standardized — no code changes
• Reusable functions used where possible instead of copy/pasted code — no code changes

General information to guide review

• Zero impact on results of default case — RPS/CES/hydrofrac values change, so results will shift in policy-binding states
• No large data file(s) added/modified
• No substantive impact on runtime for full-US reference case
• No substantive impact on folder size for full-US reference case
• No change to process flow (runreeds.py, reeds/core/solve/solve.py)
• No change to code organization
• No change to package requirements (environment.yml or Project.toml)

Did you use LLM tools (chatbot or copilot) in the preparation of this PR? If so, describe how

GitHub Copilot (chat) was used in the companion ReEDS_Input_Processing#11 for editing data_processing.py to handle the new LBNL sheet layouts, drafting the README's annual-update procedure, writing generate_comparison_plots.py, extracting text from the LBNL summary-slides PDF, and locating the v2024 NLR file. All code/text was reviewed and regenerated CSVs were spot-checked against the source workbooks before committing. No LLM usage in this repo beyond drafting this PR description.

Tag points of contact here if you would like additional review of the relevant parts of the model

• State policies: @wesleyjcole or @aschleif