Research outline

Building levees that are resistant to heavy rain disasters

fig.1 How a levee breaks
fig.1 : How a levee breaks


Photo 1 : Seepage failure of Shibui River (Miyagi Prefecture) due to heavy rain in September 2015 (break also occurred in places where there was large failure)
Photo 1 : Seepage failure of Shibui River
(Miyagi Prefecture) due to heavy rain
in September 2015
(break also occurred in places
where there was large failure)


Photo 2 : Large model experiment for seepage failure
Photo 2 : Large model experiment
for seepage failure

  Recently, major levee damage has been occurring due to typhoons and localized downpours, etc. Floods such as that of Kinugawa River during the heavy rains in Kanto and Tohoku in September 2015 are still fresh in the memory. Reinforcement of levees to guard against such river flooding is a big challenge.

  A levee is constructed over a long distance on both sides of a river, from the lower reaches to the upper reaches of the river. In Japan today, levees managed by the Nation, prefectures, and ordinance-designated cities cover approximately 73,000km (the circumference of Earth is about 40,000km). Since it would be very difficult to reinforce such a length of levees all at once, it is important to carefully inspect the levees, precisely examine levees that require reinforcement, establish an order of precedence, and reinforce levees efficiently.

  There are largely three different ways in which a levee collapses: "overtopping", in which the water level of a river exceeds the levee height and river water flows outside the levee; "erosion", in which river water washes away the earth of the levee; and "seepage failure", in which river water seeps through the inside of the levee or the ground beneath the levee and weakens the levee itself, causing it to fail (fig.1 and photo 1).

  The earth structure laboratory of the PWRI has equipment for pouring heavy rain onto a large levee model and experimental equipment that can simulate a situation in which the water level of a river has risen. Research is being carried out into structures of ways in which levees break and the extent to which levees can be efficiently reinforced, etc. through various experiments using levee models that are close to actual size, such as the above-mentioned large experiment facilities.

  Photo 2 shows an experiment that was conducted for seepage failure, the phenomenon of which at present is not adequately understood. A levee model 3m high, 6m wide and 6.5m long was produced, and when the water level of the water supply cistern was raised to simulate a rising river water level, the toe of the levee became weak, with cracks gradually appearing, and little by little failure progressed in the direction of the top of the levee. The structure of levee seepage failure was researched through experiments such as this. Going forwards, we will proceed with research into appropriate levee design methods taking into account such seepage failure, methods of preventing seepage failure, and will aim to build levees that are more efficiently resistant to heavy rain disasters.


(Contact: Soil Mechanics and Dynamics Research Team)


Introduction of research: New district-specific flood risk assessment techniques using RRI model -"Flood diagnostic chart" and "Flood hotspots"-

Table 1 Eight assessment axes in flood vulnerability assessment of each district
Table 1 Eight assessment axes in
flood vulnerability assessment of
each district



Table 2 Example of Flood Diagnostic Chart (Yazawa District)
Table 2 Example of Flood Diagnostic Chart
(Yazawa District)


Fig.1 Categorization by comprehensive scoring of each district (Red districts are Flood Hotspots)
Fig.1 Categorization by comprehensive
scoring of each district
(Red districts are Flood Hotspots)


Photo 1 Meeting for exchange of views between community leaders and disaster prevention manager from Aga Town
Photo 1 Meeting for exchange of
views between community leaders
and disaster prevention manager
from Aga Town

  ICHARM has been developing the Rainfall-Runoff-Inundation model (hereinafter, "RRI model") to research various flood risk analysis and risk management techniques.

  As part of this, since FY2014 research has been carried out to develop new flood risk analysis and risk mitigation techniques using the RRI model in order to support effective and efficient disaster prevention activities by municipalities in mountainous regions, targeting Aga Town located in the mid-stream region of Agano River of Niigata Prefecture. Specifically, flood analysis was performed with the RRI model by inputting several rainfall patterns, which include one which caused severe flood in 2011 and one which corresponds to the assumed maximum external force. Based on the results of this analysis, the flood vulnerability of each district was assessed according to 8 assessment axes, including maximum inundation height, flooding period, transport interruption and maximum number person left stranded by inundation, together with ranking using established thresholds as shown in Table 1, and scoring as risk subtotal values for each external force. This table forms the basis of the Flood diagnostic chart (see Table 2) showing assessment results in each assessment axis for each external force in each district. Comprehensive scores were then calculated by further totalling these risk subtotal values, and the 20 districts were categorized based on these comprehensive scores. As a result, particularly high-scoring districts were extracted as flood hotspots (see Fig. 1).

  The term "flood diagnostic chart" is used for the flood risk assessment technique proposed by this research, which includes the meaning of flood danger diagnosis in order to clarify the perspective from which fragility occurs districts in relation to various types of flood external force. Using the diagnosis results from this Flood diagnostic chart can be helpful not only in specifying flood hotspots, but also in drafting district disaster prevention plans in accordance with the flood characteristics of each district.

  This research is being implemented while carrying out exchanges of views with Aga Town's disaster prevention managers and community leaders as well as Niigata Prefecture and the MLIT. In March 2017, a meeting was held for an exchange of views between the community leaders of districts that were deemed to be flood hotspots and the Aga Town disaster prevention manager, and it was confirmed that the results were appropriate (see Photo 1). Going forwards, the plan is to implement initiatives for drafting district disaster prevention plans by using flood risk assessment results of each district from the flood diagnostic chart, and to research application of real-time flood analysis results using rainfall prediction.


(Contact: ICHARM)


A Study on the Development of Non-chloride Anti-icing Agents

Photo.1 Sodium propionate
Photo.1 Sodium propionate


Fig.1 Example of test results of ice-melting
Fig.1 Example of test results of ice-melting


Fig.2 Results of the metal corrosion test
Fig.2 Results of the metal corrosion test


Fig.2 A cultivation experiment result
Photo.2 A cultivation experiment result

  Ensuring road traffic safety and reliability in winter is an important social challenge in cold, snowy regions. Road administrators have implemented measures that focus on preventing the road surface from freezing, such as the application of anti-icing agents. Sodium chloride (salt) is the main anti-icing agent because of its ice-melting properties and low cost. However, concerns have been raised about the adverse impacts of years of application to bridges and other road structures. Consequently, the development of anti-icing agents that impose less of a load on bridges and other road structures has come to be an important challenge.


  The Traffic Engineering Research Team studied materials with the potential for use as novel anti-icing agents. We focused on sodium propionate(Photo.1), which is mostly used as a food additive, and we are conducting a study on its feasibility.


  To determine the basic properties of sodium propionate, tests have been conducted on whether sodium propionate meets the criteria for hazardous substance content, ice-melting properties, corrosion behavior, anti-skidding effect on roads, and effects on plants. Results showed the following: Sodium propionate met the criteria for hazardous substances, its ice-melting properties are such that it can melt ice faster than sodium chloride does, and similar to sodium chloride, its ability to melt ice decreases with decrease in temperature(Fig.1). It was also found that sodium chloride and sodium propionate mixed at a weight ratio of 8:2 is much less corrosive than sodium chloride(Fig.2), and the mixture can also improve skid resistance values on the road surface to almost the level achieved by sodium chloride. In a cultivation experiment to examine effects on plants, a mixture of sodium chloride and sodium propionate was found to have less of an adverse effect on the germination and growth of Japanese mustard spinach than sodium chloride does(Photo.2).


  In the future, towards the practical application and dissemination of the mixture of sodium chloride and sodium propionate as an anti-icing agent, verification will be performed from many different perspectives, and applicability assuming application in real situations will be reviewed.


(Contact: Traffic Engineering Research Team, Civil Engineering Research Institute for Cold Region)