A new spatiotemporal fusion model for integrating VIIRS and SDGSAT-1 Nighttime light data to generate daily SDGSAT-1 like observations

A new spatiotemporal fusion model for integrating VIIRS and SDGSAT-1 Nighttime light data to generate daily SDGSAT-1 like observations

Nighttime light (NTL) data is a critical indicator for understanding social and environmental dynamics, offering unique insights into human activities after dark. However, while providing high temporal resolution, existing NTL datasets like VIIRS suffer from low spatial resolution, limiting their ca...

Full description

Saved in:
Bibliographic Details
Main Authors: Jinhu Bian, Touseef Ahmad Khan, Ainong Li, Jinping Zhao, Yi Deng, Guangbin Lei, Zhengjian Zhang, Xi Nan, Amin Naboureh, Muhib Ullah Khan
Format: Article
Language:English
Published: Taylor & Francis Group 2025-08-01
Series:International Journal of Digital Earth
Subjects:
Data fusion
VIIRS
SDGSAT-1
nighttime light
NTLSTF model
Online Access:https://www.tandfonline.com/doi/10.1080/17538947.2025.2472912
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Nighttime light (NTL) data is a critical indicator for understanding social and environmental dynamics, offering unique insights into human activities after dark. However, while providing high temporal resolution, existing NTL datasets like VIIRS suffer from low spatial resolution, limiting their capability for detailed monitoring. There is a growing demand for NTL data with high spatial and temporal resolutions (HSTR). This study proposed a new HSTR NTL data fusion model named Nighttime Light Spatiotemporal Fusion (NTLSTF). This model generated HSTR NTL radiance values similar to SDGSAT-1 by reconstructing NTL features using a combination of spectral, spatial, and temporal weighting from VIIRS and SDGSAT-1 NTL data. Results demonstrated that the predicted SDGSAT-1 images were consistent with real SDGSAT-1 observations from both visual effect and radiance prediction accuracy. The validation of results was further supported by a high Structural Similarity Index (SSIM) of 0.976, with other quantitative metrics such as Coefficient of Determination (R²), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE), with the values of 0.941, 7.701, and 17.171, respectively. The predicted daily SDGSAT-1-like NTL data for flood disaster emergency response case in Pakistan showed the application potential of the proposed model.
ISSN:1753-8947
1753-8955

Similar Items