Medium- to high-resolution imagery is indispensable for various applications. Combining images from Landsat 8 and Sentinel-2 can improve the accuracy of observing dynamic changes on the Earth’s surface. Many researchers use Sentinel-2 10 m resolution data in conjunction with Landsat 8 30 m resolution data to generate 10 m resolution data series. However, current fusion techniques have some algorithmic weaknesses, such as simple processing of coarse or fine images, which fail to extract image features to the fullest extent, especially in rapidly changing land cover areas. Facing the aforementioned limitations, we proposed a multiscale and attention mechanism-based residual spatiotemporal fusion network (MARSTFN) that utilizes Sentinel-2 10 m resolution data and Landsat 8 15 m resolution data as auxiliary data to upgrade Landsat 8 30 m resolution data to 10 m resolution. In this network, we utilized multiscale and attention mechanisms to extract features from coarse and fine images separately. Subsequently, the features outputted from all input branches are combined and further feature information is extracted through residual networks and skip connections. Finally, the features obtained from the residual network are merged with the feature information of the coarsely processed images from the multiscale mechanism to generate accurate prediction images. To assess the efficacy of our model, we compared it with existing models on two datasets. Results demonstrated that our fusion model outperformed baseline methods across various evaluation indicators, highlighting its ability to integrate Sentinel-2 and Landsat 8 data to produce 10 m resolution data.
11 October 2023 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Monitoring ice/snow surface temperature (IST) variations with high spatial and temporal resolution data from satellites are essential for research on the mass balance of the Greenland ice sheet (GrIS). However, the tradeoff between satellite sensors' bandwidth and re-entry cycle, coupled with the influence of cloudy weather, limits their ability to fine-monitor IST. Spatiotemporal data fusion is a way of producing high spatiotemporal datasets. This study uses four spatiotemporal fusion algorithms to fuse the Landsat 8 IST data and the Moderate Resolution Imaging Spectrometer IST to generate fine spatial-temporal IST in the GrIS regions. The quantitative evaluation of the different fusion data shows that the R² are all above 0.9. The spatial and temporal nonlocal filter based fusion model (STNLFFM) dual-temporal algorithm provided the highest accuracy with a root mean square error of 2.427 K, followed by the STNLFFM mono-temporal algorithm, the spatial and temporal adaptive reflectance fusion model (STARFM), the flexible spatiotemporal data fusion model (FADSF), and enhanced STARFM (ESTARFM). From the results, the fusion data are accurate and detailed in different regions. That is, the spatiotemporal fusion technique has the potential to generate IST datasets that possess high spatial and temporal resolutions for Greenland.
Assessing the regional mass balance of European glaciers presents significant challenges due to limited measurements. While various albedo methods have been explored for individual glaciers, a comprehensive analysis of aggregated albedo methods is lacking. Addressing this gap, in our study, we examined five MODIS aggregated albedos (raw average, minimum average, average minimum, interpolated average, and cumulative) versus the annual mass balance for 12 Austrian and Norwegian glaciers from 2001 to 2020 to establish connections between them. We find that the raw average albedo is strongly correlated with the annual mass balance of Austrian glaciers (r = 0.91), while the interpolated average albedo is significantly correlated with the annual mass balance of Norwegian glaciers (r = 0.90). Moreover, we observe that high-elevation glaciers experience fewer cloud cover days, allowing the raw average albedo to reliably estimate the annual mass balance, whereas low-elevation glaciers are often obscured by clouds, potentially masking the true minimum albedo. Additionally, traditional indicators, such as the equilibrium-line altitude and accumulation area ratio, exhibit significant correlations with the annual mass balance of Norwegian and Austrian glaciers (r = 0.90 and 0.87, respectively), yet albedo demonstrates higher robustness. These findings provide a reference for selecting appropriate aggregation methods to reconstruct glacier mass balance from albedo observations.