LAD-Enhancer: A Lightweight All in One Aerial Detection Enhancer Under Adverse Weather

We propose a novel spatiotemporally self-supervised fusion method to establish long-term daily seamless global XCO₂ and XCH₄ products. Results show that the proposed method achieves a satisfactory accuracy that distinctly exceeds that of CAMS-EGG4 and is superior or close to those of GOSAT and OCO-2. In particular, our fusion method can effectively correct the large biases in CAMS-EGG4 due to the issues from assimilation data, such as the unadjusted anthropogenic emission for COVID-19.

The semi-empirical physical approach derives PM_2.5 with strong physical significance. However, due to the complex optical characteristic, the physical parameters are difficult to express accurately. Thus, combining the atmospheric physical mechanism and machine learning, we propose an optimized model. It creatively embeds the random forest model into the physical PM_2.5 remote sensing approach to simulate a physical parameter. Our method shows great optimized performance in the validations.

Compared to previous seamless global daily soil moisture (SGD-SM 1.0) products, SGD-SM 2.0 enlarges the temporal scope from 2002 to 2022. By fusing auxiliary precipitation information with the long short-term memory convolutional neural network (LSTM-CNN) model, SGD-SM 2.0 can consider sudden extreme weather conditions for 1 d in global daily soil moisture products and is significant for full-coverage global daily hydrologic monitoring, rather than averaging monthly–quarterly–yearly results.

Snow variation is a vital factor in global climate change. Satellite-based approaches are effective for large-scale environmental monitoring. Nevertheless, the high cloud fraction seriously impedes the remote-sensed investigation. Therefore, a recent 20-year cloud-free snow cover collection in China is generated for the first time. This collection can serve as a basic dataset for hydrological and climatic modeling to explore various critical environmental issues.