Large-sample evaluation of radar rainfall nowcasting for flood early warning

Risk mitigation for rainfall-triggered shallow slides and debris flows at regional scale is challenging. Early warning systems are a helpful tool to depict the location and time of future landslide events so that emergency managers can act in advance. Recently, some of the regions that are recurrently affected by rainfall triggered landslides have developed operational landslide early warning systems (LEWS). However, there are still many territories where this phenomenon also represents an important hazard and lack this kind of risk mitigation strategy.The main objective of this work is to study the feasibility to apply a regional scale LEWS, which was originally designed to work over Catalonia (Spain), to run in other regions. To do so we have set up the LEWS to Canton of Bern (Switzerland).The LEWS combines susceptibility maps to determine landslide prone areas and in real time high-resolution radar rainfall observations and forecasts. The output is a qualitative warning level map with a resolution of 30 m.Susceptibility maps have been derived using a simple fuzzy logic methodology that combines the terrain slope angle, and land use and land cover (LULC) information. The required input parameters have been obtained from regional, pan-European and global datasets.Rainfall inputs have been retrieved from both regional weather radar networks, and the OPERA pan-European radar composite. A set of global rainfall intensity-duration data has been used to asses if a rainfall event has the potential of triggering a landslide event.The LEWS has been run in the region of Catalonia and Canton of Bern for specific rainfall events that triggered important landslides. Finally, the LEWS performance in Catalonia has been assessed.Results in Catalonia show that the LEWS performance strongly depends on the quality of both the susceptibility maps and rainfall data. However, in both regions the LEWS is generally able to issue warnings for most of the analysed landslide events.

Download Full-text

Improving Spatial Rainfall Estimates at Mt. Merapi Area Using Radar-Rain Gauge Conditional Merging

Journal of Disaster Research ◽

10.20965/jdr.2019.p0069 ◽

2019 ◽

Vol 14 (1) ◽

pp. 69-79 ◽

Cited By ~ 1

Author(s):

Roby Hambali ◽

Djoko Legono ◽

Rachmad Jayadi ◽

Satoru Oishi ◽

◽

...

Keyword(s):

Temporal Resolution ◽

Early Warning ◽

Interpolation Method ◽

Rain Gauge ◽

Value Distribution ◽

Rainfall Data ◽

Warning Systems ◽

Radar Rainfall ◽

X Band ◽

Rainfall Estimates

Rainfall monitoring is important for providing early warning of lahar flow around Mt. Merapi. The X-band multi-parameter radar developed to support these warning systems provides rainfall information with high spatial and temporal resolution. However, this method underestimates the rainfall compared with rain gauge measurements. Herein, we performed conditional radar-rain gauge merging to obtain the optimal rainfall value distribution. By using the cokriging interpolation method, kriged gauge rainfall, and kriged radar rainfall data were obtained, which were then combined with radar rainfall data to yield the adjusted spatial rainfall. Radar-rain gauge conditional merging with cokriging interpolation provided reasonably well-adjusted spatial rainfall pattern.

Download Full-text

Deep Transfer Learning Approach for Identifying Slope Surface Cracks

Applied Sciences ◽

10.3390/app112311193 ◽

2021 ◽

Vol 11 (23) ◽

pp. 11193

Author(s):

Yuting Yang ◽

Gang Mei

Keyword(s):

Transfer Learning ◽

Early Warning ◽

Small Sample ◽

Crack Identification ◽

Learning Approach ◽

Surface Cracks ◽

Rock Masses ◽

Slope Surface ◽

Large Sample ◽

Monitoring And Early Warning

Geohazards such as landslides, which are often accompanied by surface cracks, have caused great harm to public safety and property. If these surface cracks could be identified in time, this would be of great significance for the monitoring and early warning of geohazards. Currently, the most common method for crack identification is manual detection, which has low efficiency and accuracy. In this paper, a deep transfer learning approach is proposed to effectively and efficiently identify slope surface cracks for the sake of fast monitoring and early warning of geohazards, such as landslides. The essential idea is to employ transfer learning by training (a) a large sample dataset of concrete cracks and (b) a small sample dataset of soil and rock masses’ cracks. In the proposed approach, (1) pretrained crack identification models are constructed based on a large sample dataset of concrete cracks; (2) refined crack identification models are further constructed based on a small sample dataset of soil and rock masses’ cracks. The proposed approach could be applied to conduct UAV surveys on high and steep slopes to provide monitoring and early warning of landslides to ensure the safety of people and property.

Download Full-text