SIFT‐Based Estimation of Zonal Drift Velocity in Equatorial Plasma Bubbles Using GOLD Observations

Abstract The Scale‐Invariant Feature Transform (SIFT) algorithm detects key points and generates descriptors in images, enabling features matching across different images for object recognition and tracking. We derived the zonal drift velocities of equatorial plasma bubbles (EPBs) by tracking SIFT key points in GOLD Nmax data. Zonal drift velocities varied from ∼40 m/s to ∼160 m/s and exhibited prominent seasonal variations: the average drift velocity for different longitudes peaks around the Northern Hemisphere's winter solstice month and reaches its minimum near the summer solstice month. The results show that the zonal drift speeds of EPBs are larger during high solar activity, while intense geomagnetic activity suppresses eastward drift velocities. This study represents the first application of the SIFT algorithm to satellite airglow images. Our findings reveal climatological variations in ionospheric zonal drifts, providing new observational foundations for advancing the understanding of ionospheric electrodynamic processes.