Introduction

In 2015, the United Nations established 17 Sustainable Development Goals (SDGs), with Goal 7 focusing on ensuring access to affordable, reliable, and sustainable modern energy for all by 2030. By 2022, approximately 760 million people, or 1 in 11 globally still lacked electricity access according to  Tracking SDG7 :The Energy Progress Report 2022, posing significant challenges to achieving this goal. Traditional survey methods for estimating the proportion of people with electricity access are often costly, infrequently updated, and hindered by the need for interpolation of historical data.

This dataset provides a 31-year harmonized global Electricity Access Indicator (EAI) from 1992 to 2022, derived from satellite nighttime light (NTL) remote sensing. Unlike conventional binary metrics that simply count nominal connections, the EAI captures functional electricity usage — measuring the proportion of the population that demonstrably uses electricity, as evidenced by detectable nighttime luminosity.

The dataset is produced by integrating two calibrated NTL archives (DMSP-CCNL, 1992–2013; NPP/VIIRS, 2012–2022) with the GlobPOP gridded population dataset. A quantile-based thresholding approach (5th percentile of non-zero NTL pixels) combined with a DMSP-to-VIIRS cross-sensor masking strategy ensures temporal consistency across the sensor transition period. The framework is validated against World Bank national statistics (R = 0.87, RMSE = 15.4%) and subnational survey data across 59 countries.

Key finding: In 2022, at least 1.06 billion people lacked functional electricity access — substantially exceeding the 685 million reported in official tracking reports, revealing a widespread prevalence of hidden energy poverty where nominal grid connections exist but functional usage remains undetectable.

Data Description

File 1 : Gridded EAI at 0.1° resolution (GeoTIFF)

Annual global rasters from 1992 to 2022. Each pixel value represents the proportion of the population with functional electricity access (0–100%, Float32) within a 0.1° × 0.1° grid cell (~11 km at the equator).

Filename convention: EAI_0dot1_Deg_WGS84_F32_{YEAR}.tif

Coverage: 180°W–180°E, 75°N–65°S

File 2 : National-scale EAI time series (CSV + Shapefile)

Country-level aggregated EAI from 1992 to 2022 for 190+ countries.

Filename: EAI_Level_0_1992_2022.csv / EAI_Level_0_1992_2022.shp

Administrative boundaries sourced from GADM v4.0.

File 3 : Electricity Accessed Population Density at 30 arc-second resolution (GeoTIFF)

Annual pixel-level maps of the population density with functional electricity access (persons/km²), at the full 30 arc-second (~1 km) resolution. Suitable for fine-scale analysis, urban–rural decomposition, or integration with other high-resolution datasets.

Filename convention: Elec_PopDen_WGS84_30arc_F32_{YEAR}.tif

Methodology Summary

Step 1  NTL harmonization.

DMSP-CCNL (30 arc-second, 1992–2013) and NPP/VIIRS (resampled to 30 arc-second, 2012–2022) are harmonized. A composite masking strategy retains DMSP-confirmed electrified pixels through the sensor transition to correct for VIIRS late-night overpass underestimation bias.

Step 2 Threshold determination.

The 5th percentile of non-zero annual NTL values (Tq = p05) is applied as the electrification threshold, selected via sensitivity analysis (p01–p10) and validated against 264,067 road-sampled ground truth points with ≥ 85% classification accuracy.

Step 3  EAI estimation.

The electrified population is computed by multiplying the binary electrified mask with GlobPOP population density. National and gridded EAI values are derived per Equation: EAI = ΣPop_light / ΣPop_total × 100%.

Step 4  Validation.

Benchmarked against World Bank and IEA national statistics (N = 5,766; R = 0.872; RMSE = 15.4%) and subnational data across 59 countries (N = 658; R = 0.825). Quasi-experimental validation using Kenya's Last Mile Connectivity Project confirms the indicator captures real-world grid densification interventions.

Applications

This dataset directly supports:

• SDG 7.1.1 monitoring : long-term functional access tracking beyond binary survey data
• Spatialized Energy Poverty Trap (SEPTs) analysis : identifying geographically entrenched stagnation zones using spatiotemporal hotspot analysis
• Policy triage : spatially distinguishing regions suitable for grid densification from those requiring off-grid decentralized deployment
• Inequality analysis : such as computing population-weighted Theil indices of subnational electrification disparities
• Electrification trajectory modelling : such as ARIMA-based projections toward 2030 SDG targets

Citation

If you use this dataset, please cite the associated paper:

Paper:

Liu, L., Chen, Y., Tanikawa, H. & Cao, X. Diagnosing spatialized energy poverty traps: Global evidence from nighttime lights. Advances in Applied Energy 22, 100275 (2026). https://doi.org/10.1016/j.adapen.2026.100275

GlobPOP (population input):

Liu, L., Cao, X., Li, S. & Jie, N. A 31-year (1990–2020) global gridded population dataset generated by cluster analysis and statistical learning. Sci Data 11, 124 (2024). https://doi.org/10.1038/s41597-024-02913-0

If you encounter any issues, please contact us via email at lulingliu@mail.bnu.edu.cn.

The source codes are available at GitHub: https://github.com/lulingliu/EAI.