Global positioning system (GPS) detection technology has several advantageous characteristics (i.e., all-weather applications, high accuracy, high spatial and temporal resolution, and low cost), and GPS tracking and monitoring techniques for water vapor have developed rapidly in recent years. The GPS-precipitable water vapor (GPS-PWV), obtained through inversion using this technology can reflect the water vapor inflow and outflow in a vertical air column above a certain area in nearly real-time, which is especially important for areas of severe water vapor variation. In this paper, we studied the relationship between GPS-PWV variation and actual precipitation. The specific aim was to identify the underlying physical mechanisms driving the variation and to further strengthen the utility of GPS-PWV in forecasts and warnings of severe convection weather. We concluded that (1) rapid rise in the GPS-PWV in the long-term low-level data predicted the arrival of precipitation and was therefore useful in weather forecasts; (2) the GPS-PWV variation was closely related with the movement of the water vapor transfer belt; (3) the atmosphere showed an unstable energy structure before the GPS-PWV increase; and (4) local motion was strongly related with the development and maintenance of precipitation.

Automatic prediction of visibility from surveillance images can provide timely warnings for transportation management departments and drivers, which is of great significance for improving the driving safety of highways in foggy weather. The currently mainstream prediction models are based on deep networks, which mainly learn visual features as clues to predict visibility levels. However, the geometric features of highways, such as the lane lines that can be observed from surveillance images are also important clues to reflect the visibility. Therefore, we propose a dual-branch fusion network driven by both geometric and visual features to achieve robust and effective visibility prediction. Specifically, we first exploit dual branches to learn geometric features of highway and deep visual features from the foggy surveillance images, respectively. We then design a fused classification module to fuse the dual-branch features to predict the visibility level. In order to simultaneously purify features during the fusion process, it utilizes a road attention block to highlight the deep visual features corresponding to the highway road area, and a lane length estimation block to extract the feature of the length of observable lane lines. Therefore, the dual-branch features can be adaptively fused to boost prediction performance. Meanwhile, we construct a real-scene foggy image dataset, which are all gathered from the surveillance video of real highways in China. We validate the effectiveness of the proposed network on this real-scene dataset and the synthetic dataset FRIDA. The experimental results show that our method can predict visibility levels more accurately than multiple existing methods.