An Improved Spatiotemporal Data Fusion Method for Snow-Covered Mountain Areas Using Snow Index and Elevation...

Area of Science:

• Earth and Planetary Sciences
• Remote Sensing
• Geosciences

Background:

• High-resolution remote sensing images are crucial for avalanche forecasting and weather studies in snow-covered regions.
• Existing spatiotemporal fusion models like ESTARFM struggle to accurately predict sudden surface changes, such as snow melt or snowfall.
• Limitations in sensor technology and atmospheric conditions hinder the acquisition of high spatial and temporal resolution images.

Purpose of the Study:

• To develop an improved spatiotemporal fusion method (iESTARFM) for snow-covered mountain areas in Nepal.
• To enhance the accuracy of similar pixel selection by simulating snow-covered changes using NDSI and DEM data.
• To generate dense time-series remote sensing images for improved monitoring of snow-covered regions.

Main Methods:

• The study introduces an improved Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (iESTARFM).
• Snow cover changes are simulated using the Normalized Difference Snow Index (NDSI) and Digital Elevation Model (DEM) data.
• Similar pixels are selected based on simulated snow cover changes, and NDSI is used to calculate weights for pixel prediction.

Main Results:

• iESTARFM effectively reduces abnormal bright patches in land areas compared to ESTARFM.
• Spectral accuracy improved with a 0.017 reduction in RMSE, a 0.013 increase in correlation coefficient (r), and a 0.013 increase in SSIM.
• Spatial accuracy improved, generating clearer textures with a 0.026 reduction in Robert's edge detection.

Conclusions:

• The iESTARFM method significantly improves prediction accuracy and maintains spatial details in remote sensing images of snow-covered areas.
• This enhanced method is valuable for generating dense time-series images crucial for avalanche forecasting and local weather studies in mountainous regions.
• The integration of NDSI and DEM data effectively addresses the limitations of previous models in capturing dynamic snow cover changes.