Near real-time rainfall monitoring at local scale is essential for urban flood risk mitigation. Previous research on precipitation visual effects supports the idea of vision-based rain sensors, but tends to be device-specific. We aimed to use different available photographing devices to develop a dense network of low-cost sensors. Using Transfer Learning with a Convolutional Neural Network, the rainfall detection was performed on single images taken in heterogeneous conditions by static or moving cameras without adjusted parameters. The chosen images encompass unconstrained verisimilar settings of the sources: Image2Weather dataset, dash-cams in the Tokyo Metropolitan area and experiments in the NIED Large-scale Rainfall Simulator. The model reached a test accuracy of 85.28% and an F1 score of 0.86. The applicability to real-world scenarios was proven with the experimentation with a pre-existing surveillance camera in Matera (Italy), obtaining an accuracy of 85.13% and an F1 score of 0.85. This model can be easily integrated into warning systems to automatically monitor the onset and end of rain-related events, exploiting pre-existing devices with a parsimonious use of economic and computational resources. The limitation is intrinsic to the outputs (detection without measurement). Future work concerns the development of a CNN based on the proposed methodology to quantify the precipitation intensity.

This paper reports the development of a very-short-range nowcast system, VIL Nowcast, which aims to provide precise forecasts of imminent rainfall, and in particular, heavy and localized events. The system is based on the vertically integrated liquid water content (VIL), which is estimated from three-dimensional radar observations as well as the 1-minute-resolution rainfall map obtained from the X-band polarimetric (multi-parameter) RAdar Information Network (XRAIN), to predict rainfall amounts over 10 minutes periods that extend to 10-60 minutes into the future. The spatial resolution of VIL Nowcast was 500 m, and nowcasts were produced at a temporal resolution of 5 minutes. Three precipitation events, of which two were isolated storms and one was a synoptic storm, were used as case studies to verify the model. The performance of VIL Nowcast was evaluated against the XRAIN radar rainfall data and an existing rainfall-rate nowcast system using the same advection scheme. The scope of the evaluation was limited mainly to the first prediction for 10 minutes ahead. It was found that VIL Nowcast showed a small, statistically significant improvement over the entire precipitation event, although its skill decreased at longer lead times and at higher thresholds. The key findings of this study are: (1) VIL Nowcast appears capable of generating skillful forecasts at short lead times, even for very localized heavy rainfall; (2) VIL Nowcast can reduce the time lag in the rainfall-rate nowcast system at initiation and peak precipitation; and (3) this system may improve the accuracy of heavy rainfall alerts provided for public activities and emergency alarms.

Recently, Japan has been hit by more frequent and severe rainstorms and floods. Typhoon Hagibis caused heavy flooding in many river basins in central and eastern Japan from 12–13 October 2019, resulting in loss of life, substantial damage, and many flood insurance claims. Considering that obtaining accurate assessments of flood situations remains a significant challenge, this study used a geographic information system (GIS)-based analytical hierarchy process (AHP) approach to develop flood susceptibility maps for the Abukuma, Naka, and Natsui River Basins during the Typhoon Hagibis event. The maps were based on population density, building density, land-use profile, distance from the river, slope, and flood inundation. A novel approach was also employed to simulate the flood inundation profiles of the river basins. In addition, a crosscheck evaluated the relationship between flood insurance claims and the developed flood risk zones within the river basins. Over 70% of insurance claims were concentrated in high to very high risk zones identified by the flood susceptibility maps. These findings demonstrate the effectiveness of this type of assessment in identifying areas that are particularly vulnerable to flood damage, which can be a useful reference for flood disaster management and related stakeholder concerns for future extreme flood events.

Urban flood risk mitigation requires fine-scale near-real-time precipitation observations that are challenging to obtain from traditional monitoring networks. Novel data and computational techniques offer a valuable potential source of information. This study explores an unprecedented, device-independent, artificial intelligence-based system for opportunistic rainfall monitoring through deep learning models that detect rainfall presence and estimate quasi-instantaneous intensity from single pictures. Preliminary results demonstrate the models’ ability to detect a significant meteorological state corroborating the potential of non-dedicated sensors for hydrometeorological monitoring in urban areas and data-scarce regions. Future research will involve further experiments and crowdsourcing, to improve accuracy and promote public resilience.

Severe storm events in recent years have caused extensive flooding in various regions of Japan, creating significant challenges for rapid emergency response and damage assessment. This study presents a real-time flood mapping approach using images and videos from social media, implemented in a browser-based web GIS tool. The system enables users to input reference point and inundation depths to estimate inundation extents based on high-resolution elevation data. We applied this tool during two recent flood events: the July 2023 flood in Akita City and the September 2024 flood in Wajima City. Notably, the latter occurred in areas still recovering from the January 2024 Noto Peninsula Earthquake, amplifying the disaster's impact on vulnerable infrastructure and temporary housing. In both cases, flood conditions were estimated within hours after impact using social media data and the results were publicly available the same day. In Wajima, field surveys were conducted for post-event validation. Although some discrepancies were observed in areas where the terrain had changed due to earthquake-induced subsidence, overall agreement between estimated and observed flood depths was reasonably good. These results indicate that real-time flood mapping using crowd-sourced data is effective for flood disaster response and can supplement existing hazard information system. The study also highlights limitations in spatial accuracy and emphasized the need for automated region segmentation and multisource data fusion to improve scalability and reliability.