With the continuous advancement of spaceborne synthetic aperture radar (SAR) technology, L-band SAR has come to play an increasingly important role in Earth observation; however, its imaging, interferometric, and polarimetric performance are susceptible to ionospheric disturbances. In equatorial nighttime data, amplitude streak artifacts induced by equatorial plasma bubbles frequently occur and have traditionally been regarded as interference factors in SAR imaging. Nevertheless, these artifacts essentially carry critical information on ionospheric irregularities and can be exploited to retrieve physical parameters such as scintillation intensity and turbulence spectral indices. Existing methods either rely on comparisons between disturbed and undisturbed images or are limited to preliminary one-dimensional parameter estimation along the track, making it difficult to achieve two-dimensional fine-scale retrieval of scintillation structures, and lacking systematic validation of their spatial correlation with equatorial ionospheric anomalies. Therefore, how to extract the two-dimensional spatial distribution of ionospheric scintillation parameters from amplitude streak artifacts and reveal their intrinsic relationship with regions of enhanced electron density has become an important scientific question in the field of ionospheric remote sensing.
In a recent study published in Space: Science & Technology, the team led by Tang Feixiang from the College of Electronic Science and Technology, National University of Defense Technology, proposed an ionospheric scintillation spectrum analysis method based on L-band SAR amplitude streaks. Grounded in phase-screen theory and the Kirchhoff diffraction model, the study establishes the theoretical relationship between the one-dimensional spectral density function of scintillation amplitude errors and irregularity parameters. Through azimuth normalized sub-band processing and digital filtering, the one-way scintillation amplitude error is extracted, followed by nonlinear least-squares fitting to retrieve the scintillation index, turbulence intensity, and spectral index. Using two consecutive data sets from the Advanced Land Observing Satellite Phased Array Type L-band SAR (ALOS PALSAR) over South America, the study generates fine-resolution two-dimensional and along-track sub-kilometer-scale distribution maps of scintillation parameters, and compares them with Global Navigation Satellite System (GNSS) total electron content (TEC) maps. The results demonstrate that SAR can provide high-resolution observations of ionospheric scintillation over thousands of kilometers within minutes, with the retrieved scintillation index showing high consistency with direct measurements, and the spectral analysis method exhibiting good detection sensitivity even for weak scintillation. The study reveals that regions with high TEC contain more Fresnel-scale small-scale irregularities, with scintillation intensity positively correlated with TEC, and that scintillation structures exhibit pronounced regional variability. This research provides an effective method for achieving high-resolution detection of equatorial ionospheric scintillation using SAR amplitude artifacts, opening a new pathway for refined understanding of the spatial structure of ionospheric irregularities.
First, this study focuses on the scientific utilization of amplitude streak artifacts in L-band spaceborne synthetic aperture radar (SAR) and systematically introduces the fundamental principles and data selection for measuring equatorial ionospheric scintillation based on these artifacts. L-band SAR is widely used in Earth observation, but its imaging quality is susceptible to ionospheric effects. One typical class of artifacts manifests as amplitude streaks that frequently appear in equatorial nighttime data and have been confirmed to be associated with equatorial plasma bubbles. Although these artifacts have negative impacts on SAR imaging and interferometry, they essentially carry critical information on ionospheric irregularities and can be exploited to retrieve physical parameters such as scintillation intensity and turbulence spectral indices. As shown in Fig. 1, the study selected two consecutive data sets acquired by the Advanced Land Observing Satellite Phased Array Type L-band SAR (ALOS PALSAR) over South America. Data Set 1 was acquired on March 16, 2011, from 01:48 to 01:54 UTC, consisting of 46 images covering approximately 2,720 km along the track. Data Set 2 was acquired on the same day from 03:29 to 03:35 UTC, consisting of 44 images covering approximately 2,600 km along the track. The two data sets are separated by about 2 hours, and both exhibit visible amplitude streaks. Fig. 2 presents the Global Navigation Satellite System (GNSS)-derived ionospheric total electron content (TEC) maps synchronized with the data acquisition times, revealing two pronounced TEC peak regions over South America, corresponding to the Equatorial Ionization Anomaly (EIA). Notably, the two satellites traversed the southern crest (Data Set 1) and the northern crest (Data Set 2), respectively, providing ideal observational conditions for investigating the relationship between scintillation and ionospheric anomalies.
Second, the study elaborates on the methodological procedure for extracting scintillation amplitude errors from amplitude streaks and performing spectral analysis to retrieve ionospheric parameters, along with the two-dimensional measurement results. Based on phase-screen theory, ionospheric irregularities can be modeled as thin screens that modulate signal phase; the free-space propagation of the signal after traversing such a phase screen can be described by Kirchhoff diffraction, giving rise to amplitude scintillation. Through azimuth normalized sub-band processing and digital filtering, the study accurately extracts the one-way scintillation amplitude error from the contaminated SAR images. The periodogram method is then employed to estimate the one-dimensional spectral density function, followed by nonlinear least-squares fitting to retrieve the turbulence intensity and spectral index, from which the one-way scintillation index is derived. As shown in Fig. 3, the two-dimensional retrieval results present the fine spatial distributions of the scintillation index, turbulence intensity, and spectral index. For Data Set 2 (the northern crest region), the scintillation index reaches up to approximately 0.5 and the logarithmic turbulence intensity reaches 35.82, indicating the presence of extremely strong ionospheric turbulence in this region. The directly measured scintillation index is highly consistent with that retrieved from spectral parameters, validating the effectiveness of the method. Data Set 1 (the southern crest region) exhibits a similar trend, with a scintillation index reaching 0.31, although the intensity is weaker than that of Data Set 2. Fig. 3 also reveals pronounced fluctuations in the scintillation parameters near the TEC peak regions, with significantly reduced scintillation intensity in the TEC trough regions, preliminarily indicating a correlation between scintillation and ionospheric anomalies.
Finally, the study presents along-track one-dimensional high-resolution measurement results and systematically analyzes the relationship between scintillation and equatorial ionospheric anomalies. In addition to two-dimensional observations, SAR data can also provide one-dimensional measurements with an along-track resolution better than 0.1 km. As shown in Fig. 4, the along-track variations of the scintillation index and turbulence intensity are consistent with the two-dimensional observations, and the retrieved scintillation index exhibits high agreement with direct measurements, with deviations not exceeding 0.04, further validating the effectiveness of the spectral analysis method. The scatter statistics in Fig. 4 reveal a positive exponential correlation between the scintillation index and turbulence intensity, consistent with theoretical expectations. By integrating the two-dimensional and one-dimensional measurement results, this study reveals the following key findings: SAR can provide fine-scale distribution maps of scintillation parameters covering over 2,700 km with a resolution better than 0.1 km within minutes; the spectral analysis method is sensitive to both very weak and moderate scintillation; scintillation intensity is positively correlated with TEC, with high-TEC regions containing more Fresnel-scale small-scale irregularities; and pronounced rapid scintillation variations with spatial scales smaller than 35 km are captured in the northern crest region. This research provides an effective method for achieving high-resolution detection of equatorial ionospheric scintillation using SAR amplitude artifacts, revealing the spatial correlation between scintillation structures and ionospheric anomalies.
