A Reproducible Computational Framework for Assessing the Spatial Performance of Climate-Sensitive Technologies

This repository provides a reproducible Python workflow to map how climate conditions affect the annual and monthly performance of climate‑sensitive technologies on a global grid. It combines (i) long‑term hourly climate “normals” derived from ERA5 (temperature and humidity) with (ii) a user‑provided technology response matrix (e.g., performance as a function of temperature and relative humidity). The result is gridded monthly and annual performance indicators that can be exported as NetCDF and GeoTIFF. The framework can be adapted to analyze a wide range of climate-sensitive technologies and infrastructure, including: Direct Air Capture (DAC) technologies Atmospheric Water Generation systems Data centers and cooling infrastructure Climate-sensitive industrial processes Water-intensive technologies Energy system performance assessments Any response matrix linking climate variables to process performance can be integrated. What this framework produces For a chosen climatological reference period (e.g., 1991–2020), the workflow generates gridded outputs (EPSG:4326) such as: annual and monthly performance totals (NetCDF) annual and monthly GeoTIFF layers for GIS workflows Indicators are technology‑dependent and defined by the response matrix (examples: CO₂ capture productivity, water use, electricity demand, heat demand). Quickstart (using the provided example) 1) Create an environment and install dependencies: pip install -r requirements.txt 2) Run the workflow Note: Preprocessing results are already provided and ready to use for different climatological reference periods (see 03_climate_normals/). Use the scripts in 01_preprocessing/ only if you need additional reference periods or variables other than temperature and humidity. In that case you must download the respective base data from ERA5 (see section “Climate Data”). Optional preprocessing (only if needed): python 01_preprocessing/era5_hourly_to_temp_rh.py python 01_preprocessing/era5_monthly_climatology.py Apply the example technology response matrix: python 02_model_application/apply_productivity_matrix.py Outputs will be written to 04_results_example/… as NetCDF and GeoTIFF. Conceptual Workflow Climate Data AggregationHourly ERA5 temperature and dew point data are aggregated into long-term hourly climatological means for selected reference periods. Derivation of Climate VariablesRelative humidity is calculated using the Magnus approximation from temperature and dew point temperature. Performance Matrix InterpolationTechnology-specific response matrices linking temperature and relative humidity to technology performance are interpolated using multidimensional linear interpolation. Spatial AggregationHourly performance values are aggregated to monthly and annual totals on a grid-cell basis. Climate Data Source ERA5(-Land) Reanalysis (ECMWF, Copernicus Climate Data Store): https://cds.climate.copernicus.eu/datasets/reanalysis-era5-land Variables used - 2 m air temperature (t2m)- 2 m dew point temperature (d2m) Spatial Resolution- Global coverage at 0.1° x 0.1° (native resolution: 9 km). Climatological reference periods 3_climate_normals/ ├── 1991_2020/ ├── 1995_2024/ └── 2005_2024/ Each folder contains average monthly hourly climatologies for temperature (°C) and relative humidity (%) on a global level for the respective reference period. This Data can be used for direct model application. Demonstration Case (DAC productivity example) The repository includes an example based on a response matrix derived from an adapted liquid-solvent direct air capture process model published by Shorey & Abdulla (2024). However, the methodology is technology-agnostic and transferable to other climate-sensitive processes. The following graph shows the resulting gridded annual gross capture rate for the demonstration case. Operational Assumptions For the DAC application, the following constraints have been applied: No operation below 0 °C Productivity capped at 40 °C Linear interpolation within the temperature–RH domain No extrapolation beyond specified matrix bounds Not covered (important limitations): altitude/pressure effects atmospheric CO₂ variability aerosol/pollution impacts Interpret outputs as first‑order climatological suitability estimates. Repository Structure requirements.txt 01_preprocessing/ ├── era5_hourly_to_temp_rh.py └── era5_monthly_climatology.py 02_model_application/ ├── apply_productivity_matrix.py ├── temp_values.csv ├── rh_values.csv └── fit_matrix_productivity.csv 03_climate_normals/ ├── 1991_2020/ ├── 1995_2024/ └── 2005_2024/ 04_results_example/ └── 1991_2020/ ├── annual_performance.nc ├── annual_performance.tif ├── *_monthly_performance.nc └── *_monthly_performance.tif Software Requirements Python ≥ 3.10 (recommended) Dependencies: numpy>=1.24xarray>=2023.1netcdf4>=1.6scipy>=1.10rioxarray>=0.15rasterio>=1.3pandas>=2.0 See [`requirements.txt`] Licensing Code (software) All source code in this repository is licensed under the MIT License (see LICENSE). Derived climate normals and example outputs (data) The derived climatological datasets in 03_climate_normals/ and the example outputs in 04_results_example/ are licensed under Creative Commons Attribution 4.0 International (CC BY 4.0) (see LICENSE-DATA). Third-party data / models (important) ERA5 / ERA5-Land (Copernicus CDS / ECMWF):This repository includes derived climatological products generated from ERA5(-Land) data obtained via the Copernicus Climate Data Store (CDS) (ECMWF). When using or redistributing these derived products, please ensure compliance with the applicable CDS/ECMWF terms and required attribution. Shorey & Abdulla (2024) DAC model (demonstration case):The DAC demonstration matrix included in 02_model_application/ is derived from an adapted version of the liquid‑solvent DAC model described by Shorey & Abdulla (2024). The original model code is available from Shorey (2024) at the referenced GitHub release. This repository does not redistribute the original or adapted third‑party source code; users should consult the upstream license/terms for reuse of that code. References Shorey, P., & Abdulla, A. (2024). Liquid solvent direct air capture’s cost and carbon dioxide removal vary with ambient environmental conditions. Communications Earth & Environment, 5(1), 607. Shorey, P. (2024). Liquid-solvent DAC (Version v1.0.0) [Computer software].https://github.com/APEX-Carleton/Liquid-solvent-DAC/releases/tag/v1.0.0 Citation If you use this dataset or framework, please cite:Wolf, P. (2026): A Reproducible Computational Framework for Assessing the Spatial Performance of Climate-Sensitive Technologies. Zenodo. DOI: 10.5281/zenodo.19604322