A high-resolution fixed-asset model can help improve the accuracy of earthquake loss assessment. We develop a grid-level fixed-asset model for China from 1951 to 2020. We first compile the provincial-level fixed asset from yearbook-related statistics. Then, this dataset is disaggregated into 1 km × 1 km grids by using multiple remote sensing data as the weight indicator. We find that the fixed-asset value increased rapidly after the 1980s and reached CNY 589.31 trillion in 2020.

Subglacial lakes are a unique environment at the bottom of ice sheets, and they have distinct features in radar echo images that allow for visual detection. In this study, we use machine learning to analyze radar reflection waveforms and identify candidate subglacial lakes. Our approach detects more lakes than known inventories and can be used to expand the subglacial lake inventory. Additionally, this analysis may also provide insights into interpreting other subglacial conditions.