population-exposure

population-exposure adds population estimates to hazard tables, polygons, and raster cells.

Install

pip install population-exposure

Python 3.11 or newer is required.

What the result estimates

assign_population() returns the estimated population represented by the selected source and reference year. Spatial hazards use coverage-weighted allocation; table hazards use exact key joins. It does not identify observed people, exact households, or who was present during an event. Boundary cells contribute according to their covered area; finer output grids redistribute source counts but do not add demographic detail. Source meaning, resolution, modeling choices, and hazard scale limit local inference.

For raster hazards, the conservation check is a numerical alignment check, not validation of the source and not an uncertainty interval. Report the source or release, population year and meaning, hazard date and threshold, allocation method, reprojection choice, incomplete support, and conservation difference when applicable.

Quick start

import pandas as pd

import population_exposure as pe

population_source = "illustrative counts"
population_year = "not applicable"
hazard = pd.DataFrame({"cell": ["A", "B"], "risk": ["high", "low"]})
population = pd.DataFrame({"cell": ["A", "B"], "population": [100, 200]})

exposed = pe.assign_population(hazard, population, cell_columns="cell")
print(f"{population_source} ({population_year}):")
print(exposed)

This toy table has no external population source or reference year. Replace it with a documented source and year for a real analysis.

Documentation

Read the full documentation for vector and raster workflows, population sources, and the API reference.

Project

Source code · Issue tracker · Contributing · Code of conduct · Changelog · MIT License