Storage and capacity rather than volume govern the water needs of United States data centers: analysis code and data

First public release of the analysis code, derived data and figures for the study "Storage and capacity rather than volume govern the water needs of United States data centers" (Zainab Ashkanani, Konstantinos Pappas, Rabi Mohtar; Texas A&M University).

Contents

• code/: the analysis pipeline, from the hourly weather fetches (NASA POWER, Open-Meteo ERA5-Land) and the cooling-model runs through the storage-yield, capacity-aware siting, grid peak-coincidence, least-cost and validation analyses, to the numbers file and the figures.
• data/inputs/: site records of the 20 IM3 siting runs, climate-cell lists, county-to-balancing-authority map, ISD station list, per-cell cooling-model summaries.
• data/results/: every derived table of the final run and numbers_v5.json, from which every reported number is read.
• data/supplementary_data/: Supplementary Data 1 to 7.
• figures/: Figures 1 to 6 and Supplementary Figures 1 to 7 as PNG (300 dpi) and EPS.

The cooling model itself is the separate repository SAAAHco/MTX_Climate_Datacenters. Raw public inputs (weather, EIA-930, ISD, EPA service areas, USGS reanalysis, census boundaries, IM3 dataset) are not included; the README lists their sources and the scripts that fetch them.

License: MIT for code; CC BY 4.0 for data and figures.