Search

Atmospheric delay correction remains a major challenge for interferometric synthetic aperture radar (InSAR) technology. In this paper, we first reviewed several commonly used methods for tropospheric delay correction in InSAR. Subsequently, considering the large volume and high temporal resolution of global navigation satellite system (GNSS) station measurement data, we proposed a method for spatial prediction of the InSAR tropospheric delay phase based on the backpropagation (BP) neural network and GNSS zenith total delay (ZTD). Using 42 Sentinel-1 interferograms over the Los Angeles area in 2021 as an example, we validated the accuracy of the BP + GNSS method in spatially predicting ZTD and compared the correction effects of BP + GNSS and five other methods on interferograms using the standard deviation (StaD) and structural similarity (SSIM). The results demonstrated that the BP + GNSS method reduced the root-mean-square error (RMSE) in spatial prediction by approximately 95.50% compared to the conventional interpolation method. After correction using the BP + GNSS method, StaD decreased in 92.86% of interferograms, with an average decrease of 52.03%, indicating significantly better correction effects than other methods. The SSIM of the BP + GNSS method was lower in mountainous and high-altitude areas with obvious terrain changes in the east and north, exhibiting excellent and stable correction performance in different seasons, particularly outperforming the GACOS method in autumn and winter. The BP + GNSS method can be employed to generate InSAR tropospheric delay maps with high temporal and spatial resolution, effectively addressing the challenge of removing InSAR tropospheric delay signals in areas with significant terrain variations.

Satellite altimetry has been proven to measure water levels in lakes and rivers effectively. The Sentinel-3A satellite is equipped with a dual-frequency synthetic aperture radar altimeter (SRAL), which allows for inland water levels to be measured with higher precision and improved spatial resolution. However, in regions at middle and high latitudes, where many lakes are covered by ice during the winter, the non-uniformity of the altimeter footprint can substantially impact the accuracy of water level estimates, resulting in abnormal readings when applying standard SRAL synthetic aperture radar (SAR) waveform retracking algorithms (retrackers). In this study, a modified method is proposed to determine the current surface type of lakes, analyzing changes in backscattering coefficients and brightness temperature. This method aligns with ground station observations and ensures consistent surface type classification. Additionally, a dual-threshold algorithm that addresses the limitations of the original bimodal algorithm by identifying multiple peaks without needing elevation correction is introduced. This innovative approach significantly enhances the precision of equivalent water level measurements for ice-covered lakes. The study retrieves and compares the water level data of nine North American lakes covered by ice from 2016-2019 using the dual-threshold and the SAMOSA-3 algorithm with in situ data. For Lake Athabasca, Cedar Lake, Great Slave Lake, Lake Winnipeg, and Lake Erie, the root mean square error (RMSE) of SAMOSA-3 is 39.58 cm, 46.18 cm, 45.75 cm, 42.64 cm, and 6.89 cm, respectively. However, the dual-threshold algorithm achieves an RMSE of 6.75 cm, 9.47 cm, 5.90 cm, 7.67 cm, and 5.01 cm, respectively, representing a decrease of 75%, 79%, 87%, 82%, and 27%, respectively, compared to SAMOSA-3. The dual-threshold algorithm can accurately estimate water levels in ice-covered lakes during winter. It offers a promising prospect for achieving long-term, continuous, and high-precision water level measurements for middle- and high-latitude lakes.

Soil moisture (SM) is a vital climate variable in the interaction process between the Earth's atmosphere and land. However, global soil moisture products from various satellite missions and land surface models are affected by inherently discontinuous observations and coarse spatial resolution, which limits their application at fine spatial scales. To address this problem, this paper integrates three diverse types of datasets from in situ, satellites, and models through Spherical cap harmonic analysis (SCHA) and Helmert variance component estimation (HVCE) to produce 1 km of spatio-temporally continuous SM products with high accuracy. First, this paper eliminates the bias between different datasets and in situ sites and resamples the datasets before data fusion. Then, multi-source SM data fusion is performed based on the SCHA and HVCE methods. Finally, this paper evaluates the fused products from three aspects, including the performance of representative sites under different climate types, the overall performance of validation sites, and the comparison with other products. The results show that the fused products have better performance than other SM products. In the representative sites, the minimal correlation coefficient (R) of the fused products is above 0.85, and the largest root mean square error (RMSE) is below 0.040 m 3 m -3 . For all validation sites, the R and RMSE of the fused products are 0.889 and 0.036 m 3 m -3 , respectively, while the R for other products is below 0.75 and the RMSE is above 0.06 m 3 m -3 . In comparison to other SM products, the fused products exhibit superior performance, generally align more closely with in situ measurements, and possess the ability to accurately and finely capture the spatial and temporal variability of surface SM.