FUT-NTL: A global dataset for future nighttime light (2025-2050) at 1 km gridded level under shared socio-economic...

Area of Science:

• Earth and Environmental Sciences
• Remote Sensing
• Urban Studies

Background:

• Nighttime light (NTL) remote sensing data is crucial for monitoring urbanization and sustainable development.
• Predicting future NTL data provides insights into human activities and environmental changes.
• Existing NTL projections lack global consistency and dynamic evolution based on Shared Socio-economic Pathways (SSPs).

Purpose of the Study:

• To generate a globally consistent, dynamically evolving future NTL dataset (FUT-NTL) from 2025 to 2050.
• To provide NTL projections under five Shared Socio-economic Pathways (SSPs) at a 1 km spatial resolution.
• To offer a valuable resource for analyzing future urbanization, socioeconomic activities, and environmental indicators.

Main Methods:

• Utilized random forest regression models for NTL prediction.
• Generated a global future NTL dataset (FUT-NTL) from 2025 to 2050 at 5-year intervals.
• Validated models against observed NTL intensity in 2020, achieving high accuracy (R² up to 0.92).

Main Results:

• The prediction models demonstrated strong global performance with high R² values and acceptable RMSE.
• Predicted NTL datasets align with SSP narratives, indicating highest growth in rapid development scenarios (especially SSP5).
• Sub-Saharan Africa exhibited the highest NTL intensity growth rates across most scenarios.

Conclusions:

• The generated FUT-NTL dataset provides a globally consistent and dynamically evolving projection of future NTL.
• The dataset is valuable for analyzing future urbanization, socioeconomic trends, and environmental impacts.
• These projections can support sustainable development planning and decision-making.