water

Case Report Lesson Learned from Catastrophic Floods in Western in 2018: Sustainable Perspective Analysis

Song-Shun Lin 1, Ning Zhang 2,3,*, Ye-Shuang Xu 1 and Takenori Hino 4

1 Department of Civil Engineering, School of Naval Architecture, Ocean, and Civil Engineering, Jiao Tong University, Shanghai 200240, ; [email protected] (S.-S.L.); [email protected] (Y.-S.X.) 2 Key Laboratory of Intelligent Manufacturing Technology (Shantou University), Ministry of Education, Shantou 515063, Guangdong, China 3 Department of Civil and Environmental Engineering, College of Engineering, Shantou University, Shantou 515063, Guangdong, China 4 Department of Civil Engineering and Architecture, Saga University, Saga 840-8502, Japan; [email protected] * Correspondence: [email protected]

 Received: 5 August 2020; Accepted: 4 September 2020; Published: 6 September 2020 

Abstract: Natural hazards have a significant impact on the sustainable development of human society. This paper reports on the catastrophic floods in western Japan in 2018. Continuous rainfall resulted in catastrophic floods, leading to 212 deaths, damage to more than 2000 houses and 619 geological disasters in 31 prefectures. The causes and contributing factors of these catastrophic floods are analyzed. The analysis of the causes of typical natural hazards provides an important lesson for hazard prevention and management. To adapt to climate change and prevent natural hazards in the future, the preliminary investigation and sustainable perspective analysis in this paper suggest the importance of the construction of a spongy city and the establishment of an early warning system with the help of information science and artificial intelligence technologies (ISAIT); we also highlight the urgent need to improve and strengthen the management of infrastructure.

Keywords: flood hazard; investigation; western Japan; early warning system; flood hazards management

1. Introduction The climate is determined by patterns of temperature, wind, , humidity and rain over a prolonged period. Global warming, caused by both natural factors and anthropic activities [1–3], is causing immediate and direct climatic change across the planet [4–8]. Flood hazards have a great influence on humans; numerous villages and farmlands are destroyed or submerged, leading a large number of people to lose their homes or lives and reducing the crop yields. Besides, traffic and communication are interrupted or blocked, and information transfer and supplies transportation are delayed. Moreover, the living standards of people in disaster-stricken areas are greatly affected. In July 2018, catastrophic floods occurred in western Japan. These catastrophic floods caused great economic losses and casualties, resulting in 212 deaths. damage to more than 2000 houses and 619 geological disasters in 31 prefectures [9]. The flood resulted in the most significant impacts on western Japan since 1982; furthermore, the number of rainfall observation stations receiving the highest amount of rainfall within 48 h compared to historical recordings was the highest since 1982. Investigations into catastrophic floods or other hazards can provide significant resources and guidelines to make comprehensive hazard prevention and management plans.

Water 2020, 12, 2489; doi:10.3390/w12092489 www.mdpi.com/journal/water Water 2020, 11, x FOR PEER REVIEW 2 of 15 Water 2020, 12, 2489 2 of 15 Investigations into catastrophic floods or other hazards can provide significant resources and guidelinesPredicting to make flood comprehensive hazards accurately hazard and prevention implementing and management effective natural plans hazard. prevention and managementPredicting plans flood can hazard mitigates accurately the impacts and from implement hazardsing and effective reduce thenatural loss tohazard the minimal prevention possible and extent.management Thus, plan it iss can necessary mitigate to the establish impacts an from early hazards warning and systemreduce (EWS).the loss Twoto the main minimal issues possible need toexten bet tackled. Thus, it in is an necessary EWS: one to isestablish data collection, an early warning and the othersystem is (EWS). data processing. Two main Dataissues concerning need to be ditackledfferent in hazards an EWS should: one is be data collected collection through, andadvanced the other technologyis data processing. and equipment. Data concerning The geographical different informationhazards should system be (GIS),collected global through positioning advanced system technology (GPS) and remoteand equipment sensing (RS). The (simplified geographical as 3S ininformation the following system context) (GIS), can global be applied positioning in data system collection; (GPS) and furthermore, remote sensing 3S has (RS) wide (simplified applications as 3S in thein the field following of hazard context prevention) can be [10 applied]. For example,in data collection Hussein; etfurthermore, al. [11] applied 3S ha remotes wide sensingapplicatio datans toin predictthe field the of occurrencehazard prevention of flash floods.[10]. For Detailed example, applications Hussein et of al RS. [11 technology] applied remote to flood sensing prediction data can to bepredict seen inthe the occurrence work of Sharma of flash et flood al. [12s. ].Detailed Zhang etapplication al. [13] utilizeds of RS sensing technology data toto mapflood coastal prediction flooding can frombe seen local in the and work global of Sharma perspectives. et al. [12 Many]. Zhang data et can al. [ be13] collected utilized sensing from 3S data technologies. to map coastal Information flooding sciencefrom local and and artificial global intelligence perspectives technologies. Many data can can provide be collected effective from means 3S oftechnologies. dealing with Information these data. Dodangehscience and et artificial al. [14] utilizedintelligen machinece technologies learning methodscan provide to predict effective flood means susceptibility. of dealing with Thus, these the use data. of 3SDodangeh technologies et al. integrated[14] utilized with machine information learning science methods and to artificial predict intelligence flood susceptibility. technologies Thus, (ISAIT) the use to constructof 3S technologies an EWS canintegrated increase with the information accuracy of hazardscienceoccurrence and artificial prediction. intelligence technologies (ISAIT) to constructThe objectives an EWS and can noveltiesincrease the of thisaccuracy paper of are hazard as follows: occurrence (1) we prediction. investigate the formation and impactsThe of objectives catastrophic and floodsnovelties in Japan; of this (2) paper we systematically are as follows analyze: (1) we theinvestigate causes of the flood; formation and (3) and we developimpacts of novel catastrophic and sustainable floods in suggestions Japan; (2) we for systematically dealing with catastrophicanalyze the causes flood hazards,of flood; and which (3) canwe providedevelop lessonsnovel and regarding sustainable countermeasures suggestions for and dealing guidelines with for catastrophic preventing flood and copinghazards with, which natural can hazards,provide lesson such ass regarding spongy city counter constructionmeasures and and the guidelines establishment for ofpreventing an early warningand coping system. with Figurenatural1 presentshazards, thesuch flowchart as spongy of city this construction article. and the establishment of an early warning system. Figure 1 presents the flowchart of this article.

Situations Impacts Prevention and Rainfall formation Casualties management Catastrophic flood Economic loss Spongy city Lesson Cause analysis Early warning system for hazards Hazards review Contributing factors Infrastructure management Current situations in Japan

Figure 1. Flowchart of the article. 2. Investigations 2. Investigations 2.1. Rainfall Formation 2.1. Rainfall Formation This section presents the formation of a catastrophic flood. Figure2 displays a specific rainfall formation;This section the area present affecteds the by catastrophicformation of flood a catastrophic in Japan is flood. also shown Figure in 2 the display figure.s a The specific affected rainfall area accountsformation for; the nearly area half affected of Japan’s by catastrophic territory. In flood July 2018, in Japan the westernis also shown regions in of the Japan figure. (see theThe red affected circle inarea Figure accounts2) were for severely nearly half a ffected of Japan by ’s territory. Prapiroon. In July 2018, The formationthe western of regions a connective of Japan airflow (see andthe thered footprintscircle in Fig ofure Typhoon 2) were Prapiroon severely are affected presented by Typhoon in the figure. Prapiroon. Heavy rainfall The formation started at of Kyushu a connective Island fromairflow 5 July and 2018 the footprints [15,16]. The of stranded Typhoon plum Prapiroon rain front are inpresented in then the crossedfigure. H Nihon-Kaieavy rainfall from started north toat southKyushu on Island 5 July from 2018 because5 July 2018 of the [15 weakening,16]. The stranded of the subtropical plum rain high-pressurefront in Hokkaido system then on crossed Pacific Ocean.Nihon-Kai Then, from the north warm to and south moist on vapor 5 July carried 2018 bybecause Typhoon of Prapiroonthe weakening crossed of thethe Pacificsubtropical Ocean. high The- informationpressure system including on Pacific the wind Ocean. speed, Then, scale the of warm wind forceand moist and the vapor tropical carried storm by grate Typhoon of five Prapiroon footprints ofcrossed Typhoon the PrapiroonPacific Ocean. is summarized The information in Table including1[ 15,16]. th Simultaneously,e wind speed, thescale plum of wind rain front—a force andffected the bytropical the high storm pressure grateon of thefive northern footprints Okhotsk of Typhoon Sea and thePrapiroon southern is Pacificsummarized Ocean—remained in Table 1 over[15,16 the]. westernSimultaneously, regions ofthe Japan, plum rain and front heavy— rainfallaffected occurred by the high in westernpressure Japanon the when northern the warmOkhotsk and Sea moist and the southern Pacific Ocean—remained over the western regions of Japan, and heavy rainfall occurred Water 20202020, 1112, x 2489 FOR PEER REVIEW 3 of 15 in western Japan when the warm and moist vapor met with the plum rain front. The continuous rainfallvapor met resulted with the in plum the heaviest rain front. floods The continuous in Japan of rainfall the previous resulted 30 in the years heaviest. The floodsformation in Japan of the of catastrophicthe previous 30 flood years. provides The formation a significant of the catastrophicresource when flood designing provides a a significant comprehensive resource hazard when preventiondesigning a plan. comprehensive hazard prevention plan.

Figure 2. FactorsFactors triggering triggering the the formation formation of of rainfall rainfall ( (datadata from [ 17]]).).

Table 1. Information for five footprints in the typhoon path. Table 1. Information for five footprints in the typhoon path. Point Tokyo Standard Time Wind Speed (m/s) Scale of Wind Force Tropical Storm Grate Point Tokyo Standard Time Wind Speed (m/s) Scale of Wind Force Tropical Storm Grate 1 13 July, 3 July,7:007:00 30 3011 11 Severe Severe tropical tropical storm storm 2 3 July, 19:00 28 10 Severe tropical storm 2 33 July, 4 July,19:00 7:00 28 2510 10 Severe Severe tropical tropical storm storm 3 44 July, 4 July, 7:00 19:00 25 2310 9Severe Tropical tropical storm storm 4 54 July, 5 July,19:00 7:00 23 209 8Tropical Tropical storm storm 5 5 July, 7:00 20 8 Tropical storm 2.2. Flooding in Western Japan 2.2. Flooding in Western Japan The western regions of Japan experienced catastrophic flooding from 5 to 9 July 2018. Of the 1300The rainfall western observation regions stationsof Japan across experienced Japan, precipitationcatastrophic atflooding 119 stations from reached5 to 9 July its highest2018. Of level the 1300within rainfall 72 h andobservation 123 stations stations received across the Japan, highest precipitation amount ofat rainfall119 stations within reached 48 h comparedits highest tolevel all withinhistorical 72 recordingsh and 123 [stations17,18]. Thereceived water the depth highest in some amount areas of was rainfall up to within 3 m, and 48 numeroush compared residual to all historicalhouses were recordings immersed [17, in18 the]. The flood. water Figure depth3 presents in som thee areas flooded was areas up to in 3 westernm, and numerous Japan. Kochi, residual Gifu, housesEhime were and Saga immerse wered thein the most flood. seriously Figure flooded3 presents areas the flooded in this catastrophic areas in western flood. Japan. The Kochi, cumulative Gifu, Ehimerainfall and of these Saga four were prefectures the most wasserious overly 600 flooded mm. Thearea floodeds in this area catastrophic of Ehime Prefectureflood. The was cumulative 970 hm2, rainfalland about of t 720hese houses four prefectures were flooded was [19 over]. As 600 shown mm. inThe Figure flooded3, the area maximum of Ehime accumulative Prefecture was rainfall 970 2 hmoccurred, and inabout Malu 720 village, houses Anji-gun, were flooded Kochi Prefecture,[19]. As show reachingn in Figure 1425.12 3, mm the inmaximum four days. accumulative The amount rainfallof rainfall occurred in many in placesMalu village, reached Anji its highest-gun, Kochi level Prefecture, in recorded reaching history. Residents1425.12 mm in Okayamain four days. County The amountreported of that rainfall floodwaters in many places arrived reached and immersed its highest their level houses in recorded in a veryhistory. short Residents time, leaving in them Countywithout reported sufficient that time flood for evacuationwaters arrived at midnight. and immersed their houses in a very short time, leaving themContinuous without sufficient heavy rainfall time for was evacuation the primary at reasonmidnight. for the flooding disasters, which were recognized as theContinuous most serious heavy flooding rainfall disasters was of the the primary previous reason 30 years. for Thethe continue flooding heavy disasters, rainfall which increased were recognizedthe rainstorm as the intensity most serious index (precipitation flooding disaster withins of thethe pastprevious 24 h). 30 The years. rainstorm The continue intensity heavy index rainfall is an increasedimportant the parameter rainstorm to analyzeintensity the index flood (precipitation in the return within period ofthe event. past 2 The4 h). return The rainstorm period of intensity recorded indexmaximum is an important rainstorm parameter intensity indices to analyze in77 the locations flood in the in Japanreturnranges period fromof event. 0.7 The to 41.8 return years period [20]. of recorded maximum rainstorm intensity indices in 77 locations in Japan ranges from 0.7 to 41.8 years [20]. To evaluate the effect of precipitation on the severity of the catastrophic flooding, a Water 2020, 12, 2489 4 of 15 WaterWater 20202020,, 1111,, xx FORFOR PEERPEER REVIEWREVIEW 44 ofof 1515

Torainstormrainstorm evaluate intensityintensity the effect indexindex of precipitation (R(R == rainfallrainfall on withinwithin the severity thethe pastpast of the2424 h)h) catastrophic isis introducedintroduced flooding, [[22]].. BasedBased a rainstorm onon thethe standardstandard intensity indexrainfallrainfall (R levellevel= rainfall [[22]],, rainfallrainfall within cancan the bebe past classifiedclassified 24 h)is intointo introduced threethree differentdifferent [2]. Based levels:levels: on (I)(I) the rrainstormainstorm standard (R(R rainfall == 5050––100100 level mm);mm); [2], rainfall(II)(II) ttorrentialorrential can be rainrain classified (R(R == 100100 into––250250 three mm);mm); di ffandanderent (II(III) levels:I) eextraordinaryxtraordinary (I) rainstorm rainstormrainstorm (R = 50–100 (R(R >> 250250 mm); mm).mm). (II) FigFig torrentialureure 44 showsshows rain (Rthethe= recordedrecorded100–250 rainfallrainfall mm); and levelslevels (III) forfor extraordinary 1111 prefecturesprefectures rainstorm fromfrom 44 toto (R 88 JulyJuly> 250 20182018 mm). (the(the Figure locationslocations4 shows ofof prefecturesprefectures the recorded havehave rainfallbeenbeen markedmarked levels forinin 11FigFig prefecturesureure 33)).. TheseThese from 11 11 4 prefectures prefectures to 8 July 2018 (Ehime, (Ehime, (the locations Fukuoka, Fukuoka, of prefectures Hiroshima, Hiroshima, have Kagoshima, Kagoshima, been marked Kochi, Kochi, in FigureKaKagawa,gawa,3). Miyazaki,Miyazaki, These 11 prefectures Nagasaki,Nagasaki, OitaOita (Ehime, ShimaneShimane Fukuoka, andand Tokushima)Tokushima) Hiroshima, Kagoshima,werewere allall floodedflooded Kochi, toto different Kagawa,different degreesdegrees Miyazaki, byby Nagasaki,thisthis flood.flood. Oita Shimane and Tokushima) were all flooded to different degrees by this flood.

FigureFigure 3.3. CumulativeCumulative rainfall rainfall distribution distribution during during 5 5–85––88 JulyJuly 20182018 (modified(modified(modified fromfrom [[2121]).]).]).

FigureFigure 4.4. SummarySummary of of the the rainfall rainfall measured measured duringduring 4–844––88 JulyJuly 2018.2018.

az 3.3. HHazardardss andand LossLoss CausedCaused byby FloodingFlooding

3.1.3.1. FloodFlood HHazardazard Flood,Flood, leveelevee breakingbreaking andand waterloggingwaterlogging werewere causedcaused byby heavyheavy rainfallrainfall andand furtherfurther triggeredtriggered secondarysecondary disasters disasters such such as as landslides, landslides, mudslides mudslides and and cliff cliff collapse collapsess.. Until Until 13 13 July July 2018, 2018, 619 619 Water 2020, 12, 2489 5 of 15

3. Hazards and Loss Caused by Flooding

3.1. Flood Hazard Flood, levee breaking and waterlogging were caused by heavy rainfall and further triggered secondary disasters such as landslides, mudslides and cliff collapses. Until 13 July 2018, 619 geological disasters occurred in 31 prefectures due to the heavy rainfall in western Japan. As shown in Table2, 168 mudslides, 427 cliff collapses and 24 landslides occurred, with cliff collapse disasters accounting for more than 60% of the total number of disasters. More than 23,000 people in 17 prefectures received refuge warnings and were forced to leave their residence. More than 2000 houses were damaged and nearly 15,000 houses were immersed under water in 31 prefectures. In total, 5074 houses collapsed during this flood hazard and 2687 houses were partially damaged. The supply of clean water and electricity in the most seriously affected areas was stopped, affecting more than 250,000 households [18]. Traffic obstacles occurred due to the disturbance of the rail and road transportation system.

Table 2. Statistics of geo-disasters (data from [18]).

Item Geological Disaster Numbers Debris flow Cliff collapsed Landslides Geo-Disaster Types 168 427 24 Hiroshima Ehime Hyogo Heaviest Affected Prefectures 123 59 51

In Kurashiki City, Okayama Prefecture, the flood water level rose rapidly, especially in low-lying areas, owing to the heavy rain. The depth of the water nearly reached the roofs of houses. Besides, in the Shinjuku-machine area of Kurashiki City, two dams on Odawa River collapsed and about 1200 ha of land were flooded [18]. Hiroshima Prefecture was most affected by this extreme natural hazard, and it suffered from the most geological disasters (see Table2). According to the work presented in [ 18], the floods triggered 97 landslides and 26 cliff collapses in Hiroshima Prefecture.

3.2. Casualties and Economic Losses Serious disasters were caused by heavy rainfall during 5–9 July 2018, leading to numerous casualties and a huge economic loss. According to the work presented in [22], the rainfall disasters resulted in 212 deaths in total, and nearly 50% of the deaths were from Hiroshima Prefecture (see Figure5). As shown in Figure5, the death toll in Hiroshima was the highest, at 101 within five days, followed by Okayama, where the death toll was 61. The majority of deaths were suffered by elderly people (70% of the total deaths), who were not capable of rescuing themselves. As listed in Table3, the economic loss to agriculture related to flooding was 16.1 billion JPY, and Hiroshima Prefecture accounted for the biggest loss of 2.5 billion JPY. Besides, according to the work presented in [22], small- and medium-sized enterprises also suffered from huge economic losses, which were estimated at 473.8 billion JPY (Okayama prefecture accounted for 281 billion JPY). It is noted that the economic loss could have been greatly underestimated, because many affected areas were not accessible. The casualties, shelters and affected areas resulting from the catastrophic flood are displayed in Figure5.

Table 3. Estimated economic loss due to the rainfall disasters during 4–9 July 2018 (Modified from [22]).

Category Economy Loss (Billion JPY) Agriculture 16.1 Public infrastructure 321 Small and medium enterprises 473.8 Forestry 23.13 Fishery 0.78 Water 2020, 12, 2489 6 of 15 Water 2020, 11, x FOR PEER REVIEW 6 of 15

Figure 5. The distribution of casualties and shelters during 4–94–9 July 2018.

4. Analysis Disasters are related to the sources of hazards and the capabilitiescapabilities of infrastructure to adapt to hazards [2121,22,22].]. Even though the sources of hazards are disastrous, flooding flooding-induced-induced hazards can be limited if the infrastructure has a strong capacity to adapt [ [23,2423,24]].. To To create create safe safe l livingiving environments, environments, increasing socialsocial investmentinvestment is is being being spent spent on on the the construction construction of of infrastructure infrastructure in in urban urban areas, area suchs, such as undergroundas underground reservoirs reservoirs [25 [25–28–],28 sewers], sewer [29s –[2932–],32 underground], underground storage storage and and drainage drainage systems system [33s –[3336–]. These36]. These underground underground infrastructures infrastructures serve serve as as reservoirs reservoirs to to store store and and regulate regulate water water underunder extreme weather conditions.

4.1. Natural H Hazardazard S Statisticstatistics over the Recent 30 YearsYears inin JapanJapan , rainstormsrainstorms and and geological geological disasters disasters caused caused by floodsby floods are referredare refer tored as to “wind as “wind and water and disasters”water disasters” in Japan. in Japan. Wind andWind water and disasterswater disasters are natural are natural hazards hazard with as strongwith a destructivestrong destructive power, leadingpower, leading to hundreds to hundreds of deaths of eachdeaths year. each Figure year.6 Figurepresents 6 present the summarys the summary of deaths of due deaths to natural due to hazardsnatural hazard from 1993s from to 2018 1993 in to Japan; 2018 thein Japan number; the of number deaths caused of deaths by catastrophic caused by catastrophic floods in 2018 flood wass the in highest2018 was of the all floodhighest hazards. of all Thisflood indicates hazards. the This insu indicatesfficiency the and insufficiency vulnerability and of the vulnerability hazard prevention of the andhazard management prevention system and management in Japan. Besides, system deathin Japan. comparisons Besides, death are conducted comparisons for are diff erentconducted hazards, for anddiffe floodrent hazards hazards, and should flood be hazards given equalshould emphasis. be given equal In Figure emphasis.6, the Hanshin-AwajiIn Figure 6, the Hanshin earthquake-Awaji in 1995earthquake (6437 deaths) in 1995 and (6437 the deaths) 2011 earthquake and the 2011 off earthquakethe Pacific coast off the of Pacific Tohoku coast in 2011 of T (22,ohok 203u in deaths) 2011 (22, are classified203 deaths) as are “Major classified Disasters” as “M inajor Japan. Disasters The flood” in Japan. in 2018 The was flood the first in 2018 time thatwas athe rainstorm first time disaster that a hadrainstorm beendesignated disaster had as been a “Special designated Emergency as a “Special Disaster” Emergency in Japanese Disaster” history in [37 Japanese]. history [37]. Water 2020, 12, 2489 7 of 15 Water 2020, 11, x FOR PEER REVIEW 7 of 15

Figure 6. Summary of missing deaths due to natural hazards from 1993 to .

4.2. Cause Analysis of Flood H Hazardsazards in Japan There are many factors that contribute to floodflood hazards.hazards. These factors are justifiedjustified based on the preliminary investigationsinvestigations of of catastrophic catastrophic floods flood ins in western western Japan Japan and and can can be classifiedbe classified into into direct direct and indirectand indirect factors. factors. The extremeThe extreme level level of accumulated of accumulated rainfall rainfall and rainfall and rainfall intensity intensity were the were direct the factors direct afactorsffecting affecting the catastrophic the catastrophic flood in 2018flood in in western 2018 in Japan. western As Japan. shown As in Figureshown4 ,in the Figure rainfall 4, inthe Hiroshima, rainfall in Kochi,Hiroshima Ehime, Kochi, and Miyazaki Ehime and reached Miyazaki over 250reached mm (extraordinaryover 250 mm (extraordinary rainstorm) in one rainstorm) day, and in the one indirect day, contributingand the indirect factors contributing were global factors warming, were a fragile global geological warming, structure, a fragile aging geological infrastructure structure and, issues aging ininfrastructure the hazards and prevention issues in and the management hazards prevention system (HPMS).and management system (HPMS). TorrentialTorrential rainsrainshave haveoccurred occurred with with a a high high frequency frequency in in recent recent years years in in Japan, Japan and, and the the frequency frequency of extremeof extreme weather weather conditions conditions is increasing. is increasing. In Figure In Fig7a,ure the 7 number(a), the of number days with of days an annual with maximuman annual temperaturemaximum temperature over 30 ◦C over per year 30 °C in per the lastyear 10 in years the last (2008–2017) 10 years was(2008 about–2017) 1.24 was times about greater 1.24 times than betweengreater than 1975 between and 1985. 1975 In and Figure 1985.7b, In the Figure annual 7(b) number, the annual of occurrences number of occurrences of rainfalls ofof morerainfalls than of 50more mm than/h in 50 the mm/ lasth 10 in years the last (2008–2017) 10 years was(2008 about–2017) 1.4 was times about that 1.4 between times 1976that between and 1985. 1976 This and indicates 1985. thatThis globalindicate warmings that global is a potential warming contributing is a potential factor contributing to catastrophic factor to floods catastrophic in western flood Japan.s in western Japan.About two-thirds of territories in Japan are mountainous, and the fragile geological structure in theAbout hilly territoriestwo-thirds showsof territories a high in liability Japan are for mountainous landslides and, and mudslides. the fragile Japan geological is a mountainous structure in islandthe hilly country; territories mountains shows a and high hills liability account for for landslides 71% of the and territory. mudslides. Japanese Japan residences is a mountainous are often island built oncountry slopes; m thatountains are easily and hills flooded account [38]. for Besides 71% of this, the mostterritory. houses Japanese in Japan residences have a wooden are often structure, built on whichslopes hasthat a are better easily resistance flooded to [38] earthquakes. Besides this but, ismost more houses easily in damaged Japan have by either a wooden flooding structure, or landslides. which Slopeshas a better and fragile resistance geological to earthquake structuress but are locatedis more neareasily residents damaged without by either proper flooding management or landslides. before extremeSlopes and heavy fragile rain; geological this will structure easily triggers are located secondary near residents disasters without under the proper influence management of heavy before rain. Concerningextreme heavy catastrophic rain; this floodwill easily hazard-a triggerffected secondary areas, Hiroshima disasters Prefectureunder the suinfluenceffered the of most heavy victims. rain. InConcer Hiroshima,ning catastrophic massive landslides flood hazard and mudslides-affected areas, caused Hiroshima by rainstorms Prefecture and flooding suffered destroyed the most victims. or even buriedIn Hiroshima, numerous massive residential landslides houses. and Tra mudslidesffic was paralyzed caused by in areasrainstorms with serious and flooding secondary destroyed geological or disasterseven buried (such numerous as soil collapse residential and debris houses. flow), Traffic which was caused paralyzed difficulties in areas for transporting with serious large secondary rescue machinerygeological disasters to the aff ected(such areasas soil to collapse save lives and within debris the flow), vital which window caused of the difficult first 72ies h. for transporting large rescue machinery to the affected areas to save lives within the vital window of the first 72 h. Water 2020, 12, 2489 8 of 15 Water 2020, 11, x FOR PEER REVIEW 8 of 15

60 400 Data from [16,17,20] Data from [16,17,20] 55 (1976-2017) 350 Linear fit 50 (a) 47 (average 1976-1985) (average 2008-2017) 45 38 300 238 (b) 40 250 35 174 30 200 25 y=0.2969x-554.58 2 150 20 R =0.2190 (1975-2017) y=1.928x-3646.95 15 100 Linear fit R2=0.1565 10 (average 1975-1985) 50

5 (average 2008-2017) Numbers of rainfall over 50 mm/hour Days with maximun temperature over 30℃ 0 0 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 Year Year Figure 7.7. ((a)a) DaysDays with aa maximummaximum temperaturetemperature over 30 ◦°CC perper yearyear ((b)b) numbersnumbers ofof rainfallrainfall eventsevents over 50 mmmm//hh annuallyannually duringduring thethe periodperiod 1975–20171975–2017 inin Japan.Japan. A large amount of infrastructure was damaged and collapsed in this catastrophic flood, including A large amount of infrastructure was damaged and collapsed in this catastrophic flood, bridges and dams. This indicates that public infrastructure is incapable of adapting to major natural including bridges and dams. This indicates that public infrastructure is incapable of adapting to hazards. Moreover, numerous bridges, reservoirs, dams and tunnels were constructed more than half a major natural hazards. Moreover, numerous bridges, reservoirs, dams and tunnels were constructed century ago in Japan; most roads in Japan were built around the 1920s. Some bridges (about 14,000) and more than half a century ago in Japan; most roads in Japan were built around the 1920s. Some bridges tunnels (about 600) were even built earlier [38] and are still in service today. At the same time, with the (about 14,000) and tunnels (about 600) were even built earlier [38] and are still in service today. At development of construction technology, the geographical environment around these bridges and the same time, with the development of construction technology, the geographical environment tunnels has been affected by human activities, making the historical infrastructure fragile to disasters. around these bridges and tunnels has been affected by human activities, making the historical Problems such as aging decay are currently reducing the resilience of the infrastructure, which needs to infrastructure fragile to disasters. Problems such as aging decay are currently reducing the resilience be renovated and updated. According to the survey results released by the Japanese Ministry of Land of the infrastructure, which needs to be renovated and updated. According to the survey results and Communications in the summer of 2017, 398 bridges and 27 tunnels, footbridges and other ancillary released by the Japanese Ministry of Land and Communications in the summer of 2017, 398 bridges facilities were assessed as “urgent measures to be taken” between 2014 and 2016—the most serious level and 27 tunnels, footbridges and other ancillary facilities were assessed as “urgent measures to be in the four-level assessment [38,39]. Since 1995, the performance of main structures, such as bridges, taken” between 2014 and 2016—the most serious level in the four-level assessment [38,39]. Since 1995, has been improved to better adapt to natural hazards. The ability—termed resilience—to withstand the performance of main structures, such as bridges, has been improved to better adapt to natural and recover from hazards is important for a city. To strengthen the resilience of city, a large hazards. The ability—termed resilience—to withstand and recover from hazards is important for a amount of work has been done, such as upgrading the hazard prevention and mitigation performance city. To strengthen the resilience of Kobe city, a large amount of work has been done, such as of facilities and increasing publicity regarding disaster prevention knowledge. However, these tasks upgrading the hazard prevention and mitigation performance of facilities and increasing publicity are not well implemented in remote areas and underdeveloped cities [40,41]. Many houses and villages regarding disaster prevention knowledge. However, these tasks are not well implemented in remote near slopes or fragile geological structures have been damaged or submerged by catastrophic floods. areas and underdeveloped cities [40,41]. Many houses and villages near slopes or fragile geological Considering the massive flood disasters in Japan, the early warning and disaster-relief mechanisms structures have been damaged or submerged by catastrophic floods. in Japan have been questioned. The Japanese government has focused on earthquake prevention, Considering the massive flood disasters in Japan, the early warning and disaster-relief which has meant that they have failed to implement sufficient countermeasures and plans for other mechanisms in Japan have been questioned. The Japanese government has focused on earthquake types of disasters, such as floods. As shown in Figure6, which presents the deaths caused by di fferent prevention, which has meant that they have failed to implement sufficient countermeasures and hazards, the Japanese government has emphasized earthquake prevention but omitted other events, plans for other types of disasters, such as floods. As shown in Figure 6, which presents the deaths which leads to high risks and insufficient countermeasures for other types of disasters, such as caused by different hazards, the Japanese government has emphasized earthquake prevention but rainstorms and flooding. The public is less aware of the methods of prevention of rainstorm disasters omitted other events, which leads to high risks and insufficient countermeasures for other types of and lacks an understanding of the criteria for special rainstorm warnings. Besides, more serious debris disasters, such as rainstorms and flooding. The public is less aware of the methods of prevention of flows, landslides and other disasters have occurred than the government and population expected. rainstorm disasters and lacks an understanding of the criteria for special rainstorm warnings. Besides, The rainstorm in 2018 covered a wide area and lasted for a long time, and the past practices, which relied more serious debris flows, landslides and other disasters have occurred than the government and partly on the adjacent areas to provide disaster relief, were ineffective, meaning that not all localities population expected. The rainstorm in 2018 covered a wide area and lasted for a long time, and the were able to take care of themselves. HPMS have been established in Japan for a long time; however, past practices, which relied partly on the adjacent areas to provide disaster relief, were ineffective, the catastrophic flood which occurred in western Japan indicated that improvements should be made meaning that not all localities were able to take care of themselves. HPMS have been established in to HPMS in Japan. Japan for a long time; however, the catastrophic flood which occurred in western Japan indicated that improvements should be made to HPMS in Japan. Water 2020, 12, 2489 9 of 15 Water 2020, 11, x FOR PEER REVIEW 9 of 15

5. SuggestionsSuggestions

5.1. Strengthen River Management and Construct Sponge C Citiesities A spongyspongy city city is anis eanffective effective method method to adapt to toadapt flood to hazards. flood Inhazards. 1980, the In Ministry 1980, the of ConstructionMinistry of (MOC)Construction in Japan (MOC) formulated in Japan a formulated set of low-impact a set of urbanlow-impa rainwaterct urban management rainwater management systems simulating systems simulatingnatural drainage, natural aiming drainage, at solving aiming the at prominent solving problemsthe prominent in the problems process of in urbanization, the process such of urbanization,as frequent waterlogging, such as frequent aggravated waterlogging, runoff aggravatedpollution, waterrunoff resource pollution, loss water and resource the deterioration loss and theof the deterioration water ecological of the water environment. ecological The environment. comprehensive The comprehensive management of management rainwater resources of rainwater was resourcescarried out was by promotingcarried out theby promoting “Rainwater the Storage “Rainwater and Infiltration Storage and Plan”. Infiltration In 1992, Plan”. the MOC In 1992, in Japan the promulgatedMOC in Japan the promulgated “Second Generation the “Second Urban Generation Sewerage Urban Master Sewerage Plan”. Master In 2002, Plan”. Japan In launched 2002, Japan the “Comprehensivelaunched the “Comprehensive Rainwater Policy-related Rainwater Policy Action-related Plan”, aimingAction toPlan”, reduce aiming urban to floods,reduce ensureurban floods, a good waterensure cycle a good in water the basin, cycle alleviate in the basin, the phenomenonalleviate the phenomenon of “heat islands” of “heat in theisland urbans” in circle the urban and guidecircle andthe formationguide the of formation the “natural” of the circulation “natural” of circula rainwatertion [42of ,43rainwater]. In recent [42,43]. years, In the recent comprehensive years, the comprehensiverainwater planning rainwater measures—named planning measures the spongy—named city scheme—have the spongy been city employed, scheme— ashave shown been in employedFigure8[ 44, as–47 show]. Then in scheme Figure related8 [44–47 to]. The a spongy scheme city related has improved to a spong flexibilityy city has in improved terms of flexibility adapting into terms environmental of adapting changes to environmental and responding changes to natural and respon hazardsding caused to natural by rain. hazards Strengthened caused by riverrain. Strengthenedmanagement isriver a critical management way to store is a andcritical drain way water to store in flood and hazardsdrain water and toin irrigateflood hazards and replenish and to waterirrigate in and drought replenish periods. water in drought periods.

Sewer regulation and Extensive outdoor Small green roofs resource recycling greening

Water Develop diversified Natural river conservation area greening water storage

Watershed Integrated rainwater Water collection Slow flow region Building dams planning planning measures facilities

Optimizing pavement Collect rainwater Low-lying areas types in many ways

Permeable Drainage road Water retaining Greening pavement surface pavement pavement

Figure 8. Comprehensive rainwater planning measures. Figure 8. Comprehensive rainwater planning measures. A suitable regional sponge city construction plan has been proposed based on local rainfall, urbanA residential suitable regional density sponge and population city construction density. planThe hasmajor been benefits proposed obtained based from on local spong rainfall,y city constructionurban residential include density the andreduced population possibility density. of Theurban major inundation, benefits obtainedthe alleviat fromion spongy of urban city constructionwaterlogging include and the the replenish reducedment possibility of groundwater. of urban inundation, The numerous the alleviation drainage of and urban storage waterlogging systems andimprove the replenishmentthe drainage and of groundwater.storage capacity The of numerous spongy cities, drainage which and could storage prevent systems some improve small-scale the drainagefloods. Regard and storageing a super capacity flood, of spongy although cities, the which amount could of water prevent might some exceed small-scale the storage floods. capacity Regarding of athe super spong flood,y city, although the spong they city amount could of still water delay might the flood exceed at the initial storage stage capacity; this would of the spongyincrease city,the thetime spongy not only city couldfor citizens still delay to thetransfer flood atto the a initialsafe place stage; but this wouldalso for increase the government the time not onlyto take for citizenscountermeasures. to transfer This to a is safe thus place benefi butcial also to foralleviate the government the hazardous to take results countermeasures. of floods. A high This drainage is thus beneficialcapacity would to alleviate also improve the hazardous the recovery results oftime floods. of cities A high from drainage the flood. capacity Besides, would the alsosilt in improve rivers theshould recovery be cleaned time ofup cities in time from to increase the flood. the Besides, volume theof water silt in storage rivers should during beflood cleaned hazards. up in Since time the to increaseimplementation the volume of the of water spongy storage city duringin 2002 flood in Tokyo, hazards. flooded Since the area implementations and houses ofdisplayed the spongy a 2 citydescend in 2002ing intrend Tokyo, from flooded 1.37 to areas 0.33 km and2 in houses 2002, displayed2004 and 2006, a descending when the trend annual from rainfall 1.37 to was 0.33 160, km 199in 2002,and 172 2004 mm and, respectively 2006, when the[48]. annual In Japan, rainfall taking was the 160, Tokyo 199 and Sky 172 Tree mm, (Japan respectively’s tallest [48tower]. In) Japan,as an takingexample the, which Tokyo is Sky an important Tree (Japan’s section tallest of tower)the drainage as an example,and storage which system is an, the important system can section store of 7000 the drainagetons of rainwater. and storage Moreover, system, an the underground system can store reservoir 7000 tonsof 4000 of rainwater.tons has been Moreover, built at anShibuya underground Station (transportationreservoir of 4000 hub tons of Tokyo). has been When built the at Shibuyarainfall exceeds Station 50 (transportation mm per hour hub, it can of store Tokyo). the Whenrainwater the in the surrounding area. At the same time, it is connected with Tokyo’s huge underground drainage system to discharge rainwater after the peak of heavy rain. Under this condition, the spongy city Water 2020, 12, 2489 10 of 15 rainfall exceeds 50 mm per hour, it can store the rainwater in the surrounding area. At the same time, it is connected with Tokyo’s huge underground drainage system to discharge rainwater after the peak of heavy rain. Under this condition, the spongy city represents a preferred way to cope with flood events. In China, many spongy cities have been applied in coastal cities, such as Shanghai, Xiamen, Ningbo and Shenzhen.

5.2. Establishment of an Early Warning System for Flood-Hazards An early warning system for flood hazards based on information science and artificial intelligence technology should be established. The prediction of hazards is a complicated systematic process because it involves many fields, such as nature, society and the economy. The different range and degree of the selected index factors for natural hazards prediction can interact with each other on the spatial scale, which results in the increased complexity of natural hazard prediction. Under these conditions, this paper develops a framework of a flood hazard early warning system. Figure9 shows the framework of an early warning system for flood hazards. This system includes four sub-systems: a monitoring and data collection system, data processing system, dynamic disaster early warning system and management system. It could be established through the Japan Meteorological Agency (JMA). In the monitoring and data collection system, the data are collected using advanced technologies: remote sensing (RS), geographic information system (GIS) and global positioning system (GPS). The collected data in the monitoring and data collection system include temperature, wind speed, rainstorm intensity index, atmospheric pressure, relative humidity and total cloud cover. The dynamic early warning system utilized these collected data to predict the precipitation. The predicted precipitation within 24 h (aforementioned rainstorm intensity index) works as the warning index for a flood hazard. The warning (hydrological) threshold is the maximum rainstorm intensity index that one city could sustain. The assigned hydrological threshold could be empirically obtained from the local return period of flood events or estimated using the capacities of local drainage and storage systems. A warning will be released when the predicted precipitation should be over the hydrological threshold. Apparently, the hydrological thresholds are distinguished for different cities, which have different drainage and storage systems. For instance, Tokyo should be assigned with a higher hydrological threshold owing to the advanced drainage and storage systems of sponge city than other cities such as saga. The data are collected by 3S technologies in a huge volume. Thus, artificial intelligence technologies (machine learning algorithms and deep learning algorithms) and big data technologies are applied to deal with data processing issues [12–14]. These new technologies have the ability to deal with data with a high non-linear correlation. For example, the multiple influential variables, such as temperature, wind speed, atmospheric pressure, relative humidity and total cloud cover are input to the system. Precipitation is a predicted variable through the early warning system as the warning index. The predicted result can provide the guideline to design the drainage and storage systems on improving the storage capacity of spongy cities. For instance, if the early warning system of one city frequently releases warnings, it implies that the warning threshold is too low. The existing drainage and storage system also could not afford the coming precipitation. One city should then improve the drainage and storage system to increase the warning threshold in the future. Besides, the flood risk can be mapped using the prediction result to provide guidelines on how to escape from the flooded area. According to the results from dynamic disaster early warning systems, the government can present a disaster prevention plan, relief plan and guidelines for the construction of infrastructure. The established warning system can be used to predict future disasters, provide relevant design principles for infrastructure construction and reinforce and transform aging infrastructure in a timely manner. It will also provide scientific decision-making support for comprehensive natural hazards risk management and emergency plans for disaster prevention [14,49–53]. For example, seasonal European early warning systems have been developed and have been running in a preoperational mode since mid-2018 under the EU-funded Enhancing Emergency Management and Response to Extreme Weather and Climate Events project [48]. Water 2020, 12, 2489 11 of 15 Water 2020, 11, x FOR PEER REVIEW 11 of 15

Flood Water 2020, 11, x FOR PEER REVIEW 11 of 15 GPS Monitoring and data RS collectionFlood system GIS Devices Historical disaster GPS casesMonitoring and data RS collection system GIS Data processing system Devices Historical disaster Big data cases Machine learning Deep learning technologies algorithms algorithms Data processing system Big data Machine learning Deep learning technologiesDynamic disasteralgorithms early warning systemalgorithms

Multi-disaster management system Dynamic disaster early warning system Disaster Disaster Guidance for construction and prevention plan relief plan reinforcement of infrastructure Multi-disaster management system Disaster Disaster Guidance for construction and FigureFigure 9. Multi9. Multiprevention disaster disaster collaborativecollaborative plan relief early early plan warning warningreinforcement syst system.em. GPS of, global GPS,infrastructure positioning global positioning system; GIS system;, GIS, geographicalgeographical information information system; system; RS, RS,remote remote sensing. sensing. Figure 9. Multi disaster collaborative early warning system. GPS, global positioning system; GIS, 5.3. Management5.3. Management of Infrastructure of Infrastructure geographical information system; RS, remote sensing. In theIn floodthe flood disaster disaster in in 2018, 2018 a, a great great deal deal ofof infrastructure,infrastructure, such such as assmall small reservoirs reservoirs and anddams dams in in Hiroshima,5.3.Hiroshima, Management Yamaguchi, Yamaguchi, of Infrastructure Okayama Okayama and and Ehime, Ehime, collapsedcollapsed due due to to aging. aging. Most Most of these of these facilities facilities were were built in the 1970s and 1980s, and they were not able to resist the impact of large-scale mountain built in theIn the 1970s flood and disaster 1980s, in 2018 and, a they great were deal of not infrastructure, able to resist such the as impactsmall reservoirs of large-scale and dams mountain in torrents and high water levels. Besides, these infrastructure projects were designed and built based torrentsHiroshima, and high Yamaguchi, water levels. Okayama Besides, and Ehime these, infrastructure collapsed due to projects aging. Most were of designed these facilities and built were based on the highest water level of the historical records at that time; as seen in Figure 7, the numbers of built in the 1970s and 1980s, and they were not able to resist the impact of large-scale mountain on therainfall highest events water over level 50 mm of the/h show historical an upward records trend. at that The time; frequent as seen heavy in rainfall Figure 7result, thes numbersin a torrents and high water levels. Besides, these infrastructure projects were designed and built based of rainfallcontinuous events increase over of 50 the mm water/h show level in an the upward reservoirs. trend. In the The catastrophic frequent flood heavy in 2018, rainfall 119 stations results in a on the highest water level of the historical records at that time; as seen in Figure 7, the numbers of continuousreached increase their highest of the level water within level 72 in h the and reservoirs. 123 stations In received the catastrophic their highest flood amount in 2018, of 119rainfall stations rainfall events over 50 mm/h show an upward trend. The frequent heavy rainfall results in a within 48 h compared to all historical recordings. The water level continuously exceeds historical reachedcontinuous their highest increase levelof the withinwater level 72 hin andthe reservoirs. 123 stations In the received catastrophic their flood highest in 2018, amount 119 stations of rainfall recordings. The continuous rainfall and record-breaking water levels not only overloaded the dams, withinreached 48 h comparedtheir highest to level all historicalwithin 72 recordings.h and 123 stations The waterreceived level the continuouslyir highest amount exceeds of rainfall historical which increased the risks of infrastructure collapse but exceeded the designed standard of water recordings.within 48 The h compared continuous to all rainfall historical and recordings. record-breaking The water water level levels continuousl not onlyy exceeds overloaded historical the dams, levels. The infrastructure then had to release flood waters and thus failed to protect cities from flood. whichrecordings. increased The the continuous risks of infrastructurerainfall and record collapse-breaking but water exceeded levels not the only designed overload standarded the dams, of water Therefore, the management and maintenance of existing infrastructure should be strengthened, and levels.which The increased infrastructure the risks then of hadinfrastructure to release collapse flood waters but exceed and thused the failed designed to protect standard cities of fromwater flood. the engineering design standards for new infrastructure in Japan should be improved. Figure 10 Therefore,levels. theThe managementinfrastructure andthen maintenancehad to release offlood existing waters infrastructure and thus failed should to protect be strengthened,cities from flood. and the presents the concept of infrastructure management from a sustainable perspective. This framework Therefore, the management and maintenance of existing infrastructure should be strengthened, and engineeringof three designstages of standards infrastructure: for new the infrastructuredesign stage, construction in Japan should stage beand improved. operational Figure stage. 10In presentsthe the engineering design standards for new infrastructure in Japan should be improved. Figure 10 the conceptinitial design of infrastructure stage, the infrastructure management is fromdesign aed sustainable based on perspective.a sustainable Thisand frameworkhazard-resistant of three presents the concept of infrastructure management from a sustainable perspective. This framework stagesperspective of infrastructure:. This should the designlower the stage, environmental construction impact stage during and operationalthe construction stage. stage. In the The initial of three stages of infrastructure: the design stage, construction stage and operational stage. In the operational stage is the most significant stage; it is composed of three systems (monitoring, evaluation designinitial stage, design the infrastructurestage, the infrastructure is designed is de basedsigned on based a sustainable on a sustainable and hazard-resistant and hazard-resistant perspective. and maintenance and repair systems) to ensure the operation of the infrastructure. Thisperspective should lower. This the should environmental lower the impact environmental during theimpact construction during the stage. construction The operational stage. The stage is the mostoperational significant stage is stage; the most it is significant composed stage of three; it is composed systems (monitoring,of three systems evaluation (monitoring and, evaluation maintenance and repairand maint systems)enance toSustainabilityand ensure repair thesystem operations) to ensure of the the infrastructure. operationExternal loadof the infrastructure. Hazards resistant

SustainabilityDesign stage InfrastructureExternal load health monitored system Hazards resistant Monitored Feedback data Technical Design stage Infrastructure health Constrcuion monitoredlevel system stage MonitoredInfrastructure health MaintenanceFeedback and evaluationdata system repair system Environmental TechnicalFeedback impactConstrcuion level stage Infrastructure health Maintenance and evaluation systemOperational stage repair system Environmental Feedback impact Figure 10. Management of infrastructureOperational based stageon a sustainable view.

Figure 10. Management of infrastructure based on a sustainable view. Figure 10. Management of infrastructure based on a sustainable view. Water 2020, 12, 2489 12 of 15

5.4. Other Recommendations Apart from the suggestions mentioned above, flood hazard awareness should be increased. For example, the population should stay away from dangerous areas (riverbeds, reservoirs and channels and culverts). The people (in dangerous areas) should leave after receiving early warning of flood hazards. Besides, additional technical flood protection measures are also important. Infrastructure, such as dams, dikes and reservoirs, should be built and maintained in time to mitigate the impacts of flood hazards. Furthermore, houses should be built with improved engineering standards, which can give houses a better adaptive ability during flood hazards.

6. Conclusions This paper analyzes the extreme flooding hazard that occurred during 5–9 July 2018 in western Japan. According to the preliminary investigation and analysis of flooding hazards, concluding remarks from a sustainable perspective are presented below.

(1) The catastrophic flood that occurred in western Japan in July 2018 led to 212 deaths. This flood hazard also led to more than 2000 houses being damaged/destroyed and 619 geological disasters in 31 prefectures. The impacts of the catastrophic flooding hazard in western Japan revealed the vulnerabilities of hazard prevention and management systems in Japan. The causes of and contributing factors to the catastrophic flood are illustrated. The analysis of the catastrophic flood in Japan provides a valuable lesson for flood hazard prediction, prevention, and management. (2) Some countermeasures are presented to better prevent and cope with flood hazards in the future. A spongy city should be constructed to enhance the resilience to adapt to extreme rainfall events. In addition, a framework for a flood hazard early warning system is proposed. A collaborative early warning system is established based on information science and artificial intelligence technologies. Finally, the importance of maintaining and updating infrastructure in time is highlighted.

Author Contributions: This paper represents a result of collaborative teamwork. T.H. and Y.-S.X. developed the concept; S.-S.L. draft the manuscript; and N.Z. provided constructive suggestions and revised for the manuscript. The four authors contributed equally to this work. All authors have read and agreed to the published version of the manuscript. Funding: The research work described herein was funded by the Research Funding of Shantou University for New Faculty Member (Grant No. NTF19024-2019). These financial supports are gratefully acknowledged. Conflicts of Interest: The authors declare no conflict of interest.

Nomenclature

Items Abbreviation Carbon dioxide CO2 Ehime Eh Fukuoka Fu Geographical information system GIS Global positioning system GPS Hiroshima Hi Japanese Yen JPY Hazards prevention and management system HPMS Information science and artificial intelligence technologies ISAIT Kagoshima Ka Kochi Ko Water 2020, 12, 2489 13 of 15

Items Abbreviation Kagawa Kag Ministry of Construction MOC Miyazaki Mi Early warning system EWS Nagasaki Na Oita Oi Remote sensing RS Shimane Sh Tokushima To

References

1. Lyu, H.M.; Shen, S.L.; Zhou, A.; Yang, J. Risk assessment of mega-city infrastructures related to land subsidence using improved trapezoidal FAHP. Sci. Total Environ. 2020, 717, 135310. [CrossRef][PubMed] 2. Lyu, H.M.; Wang, G.F.; Shen, J.S.; Lu, L.H.; Wang, G.Q. Analysis and GIS mapping of flooding hazards on 10 May 2016, Guangzhou, China. Water 2016, 8, 447. [CrossRef] 3. Lyu, H.M.; Shen, S.L.; Zhou, A.; Zhou, W.H. Data in flood risk assessment of metro systems in a subsiding environment using the interval FAHP-FCA approach. Data Brief 2019, 26, 104468. [CrossRef][PubMed] 4. Wu, Y.X.; Lyu, H.M.; Shen, S.L.; Zhou, A. A three-dimensional fluid-solid coupled numerical modeling of the barrier leakage below the excavation surface due to dewatering. Hydrogeol. J. 2020, 28, 1449–1463. [CrossRef] 5. Wu, Y.X.; Shen, S.L.; Lyu, H.M.; Zhou, A.N. Analyses of leakage effect of waterproof curtain during excavation dewatering. J. Hydrol. 2020, 583, 124582. [CrossRef] 6. Intergovernmental Panel on Climate Change (IPCC). Climate Change: Radiative Forcing of Climate Change and an Evaluation of the IPCC IS92 Emission Scenarios; Cambridge University Press: Cambridge, UK, 1995. 7. Lin, S.S.; Shen, S.L.; Zhou, A.; Lyu, H.M. Sustainable development and environmental restoration in Lake Erhai, China. J. Clean. Prod. 2020, 258, 120758. [CrossRef] 8. Ward, P.J.; Kummu, M.; Lall, U. Flood frequencies and durations and their response to El Niño Southern Oscillation: Global analysis. J. Hydrol. 2016, 539, 358–378. [CrossRef] 9. Ali, R.S.; Kuriqi, A.; Kisi, O. Human-environment natural disasters interconnection in China: A review. Climate 2020, 8, 48. [CrossRef] 10. Cabinet Office. (COGOJ). Damage Caused by Heavy Rain on July 9th, 30. Available online: http://www.bousai.go.jp/updates/h30typhoon7/index.html (accessed on 5 May 2019). (In Japanese) 11. Hussein, S.; Abdelkareem, M.; Hussein, R.; Askalany, M. Using remote sensing data for predicting potential areas to flash flood hazards and water resources. Remote Sens. Appl. Soc. Environ. 2019, 16, 100254. [CrossRef] 12. Sharma, T.P.P.; Zhang, J.H.; Koju, U.A.; Zhang, S.; Bai, Y.; Suwal, M.K. Review of flood disaster studies in Nepal: A remote sensing perspective. Int. J. Disaster Risk Reduct. 2019, 34, 18–27. [CrossRef] 13. Zhang, K.Q.; Gann, D.; Ross, M.; Robertson, Q. Accuracy assessment of ASTER, SRTM, ALOS, and TDX DEMs for Hispaniola and implications for mapping vulnerability to coastal flooding. Remote Sens. Environ. 2019, 225, 290–306. [CrossRef] 14. Dodangeh, E.; Choubin, B.; Eigdir, A.N.; Nabipour, N.; Panahi, M.; Shamshirband, S.; Mosavi, A. Integrated machine learning methods with resampling algorithms for flood susceptibility prediction. Sci. Total Environ. 2020, 705, 135983. [CrossRef][PubMed] 15. China National Meteorological Center. Available online: http://data.cma.cn/ (accessed on 25 July 2018). 16. Cabinet Office. Government of Japan (COGOJ). The Rainstorm Disaster in Northern Kyushu (TRDNK). 2018. Available online: http://www.bousai.go.jp/kaigirep/hakusho/h30.html (accessed on 5 May 2019). (In Japanese) 17. Japan Meteorological Agency (JMA). Available online: http://www.data.jma.go.jp/cpdinfo/extreme/extreme_ p.html (accessed on 25 July 2018). 18. Ministry of Land, Infrastructure, Transport and Tourism (MLITT). 2018. Available online: http://www.mlit. go.jp/index.html (accessed on 5 May 2019). (In Japanese) Water 2020, 12, 2489 14 of 15

19. Jiang, F.R. Japan’s catastrophic flood in July 2018 and its response. China Flood Drought Manag. 2018, 28, 12. [CrossRef] 20. Edogawa River Office. Metropolitan Area Outer Canal. Available online: https://www.ktr.mlit.go.jp/ edogawa/gaikaku/hans/edogawa_hans00823.html (accessed on 5 May 2019). (In Japanese) 21. Japan Meteorological Agency (JMA). Heavy Rain Due to Plum Rain Plum Rain Front and Typhoon Prapiroon. Available online: http://www.data.jma.go.jp/gmd/cpd/longfcst/ (accessed on 5 May 2019). (In Japanese) 22. Cabinet Office. Government of Japan (COGOJ). July Heisei Heavy Rain Emergency Disaster Countermeasure Headquarters Meeting. Available online: http://www.bousai.go.jp/updates/h30typhoon7/index.html (accessed on 5 May 2019). (In Japanese) 23. Lyu, H.M.; Sun, W.J.; Shen, S.L.; Arulrajah, A. Flood risk assessment in metro systems of mega-cities using a GIS-based modeling approach. Sci. Total Environ. 2018, 626, 1012–1025. [CrossRef] 24. Lyu, H.M.; Xu, Y.S.; Cheng, W.C.; Arulrajah, A. Flooding hazards across southern China and prospective sustainability measures. Sustainability 2018, 10, 1682. [CrossRef] 25. Lyu, H.M.; Shen, S.L.; Zhou, A.N.; Yang, J. Perspectives for flood risk assessment and management for mega-city metro system. Tunn. Undergr. Space Technol. 2019, 84, 31–44. [CrossRef] 26. Lyu, H.M.; Zhou, W.H.; Shen, S.L.; Zhou, A.N. Inundation risk assessment of metro system using AHP and TFN-AHP in Shenzhen. Sustain. Cities Soc. 2020, 56, 102103. [CrossRef] 27. Lyu, H.M.; Shen, S.L.; Yang, J.; Yin, Z.Y. Inundation analysis of metro systems with the storm water management model incorporated into a geographical information system: A case study in Shanghai. Hydrol. Earth Syst. Sci. 2019, 23, 4293–4307. [CrossRef] 28. Marchi, L.; Borga, M.; Preciso, E.; Gaume, E. Characterization of selected extreme flash floods in Europe and implications for flood risk management. J. Hydrol. 2010, 394, 118–133. [CrossRef] 29. Gallina, V.; Torresan, S.; Critto, A. A review of multi-risk methodologies for natural hazards: Consequences and challenges for a climate change impact assessment. J. Environ. Manag. 2016, 168, 123–132. [CrossRef] 30. Zhao, J.W.; Peng, F.L.; Wang, T.Q. Advances in master planning of urban underground space (UUS) in China. Tunn. Undergr. Space Technol. 2016, 55, 290–307. [CrossRef] 31. Peng, J.; Peng, F.L. A GIS-Based evaluation method of underground space resource for urban spatial planning: Part 1 Methodology. Tunn. Undergr. Space Technol. 2018, 74, 82–95. [CrossRef] 32. Qiao, Y.K.; Peng, F.L.; Wang, Y. Monetary valuation of urban underground space: A critical issue for the decision-making of urban underground space development. Land Use Policy 2017, 69, 12–24. [CrossRef] 33. Shen, S.L.; Atangana Njock, P.G.; Zhou, A.; Lyu, H.M. Dynamic prediction of jet grout column diameter in soft soil by using Bi-LSTM deep learning. Acta Geotech. 2020.[CrossRef] 34. Wang, X.W.; Yang, T.L.; Xu, Y.S.; Shen, S.L. Evaluation of optimized depth of waterproof curtain to mitigate negative impacts during dewatering. J. Hydrol. 2019, 577, 123969. [CrossRef] 35. Wu, H.N.; Shen, S.L.; Chen, R.P.; Zhou, A.N. Three-dimensional numerical modelling on localised leakage in segmental lining of shield tunnels. Comput. Geotech. 2020, 122, 103549. [CrossRef] 36. Elbaz, K.; Shen, S.L.; Sun, W.J.; Yin, Z.Y.; Zhou, A.N. Prediction model of shield performance during tunneling via incorporating improved Particle Swarm Optimization into ANFIS. IEEE Access 2020, 8, 39659–39671. [CrossRef] 37. Shen, S.L.; Wu, Y.X.; Misra, A. Calculation of head difference at two sides of a cut-off barrier during excavation dewatering. Comput. Geotech. 2017, 91, 192–202. [CrossRef] 38. Cabinet Office. Government of Japan (COGOJ). Japan’s Heisei 30 years Disaster Prevention White Paper. Available online: http://www.bousai.go.jp/kaigirep/hakusho/h30.html (accessed on 5 May 2019). (In Japanese) 39. China Research and Communication Center (CRCC). 2018. Available online: http://www.jst.go.jp/crcc/ (accessed on 5 May 2019). 40. Road Branch, Ministry of Land, Infrastructure, Transport and Tourism (MLIT). Annual Report of Road Maintenance, the Ministry of Land, Infrastructure, Transport. and Tourism; MLIT: Tokyo, Japan, 2017; 47p. 41. Lamond, J.E.; Proverbs, D.G. Resilience to flooding: Lessons from international comparison. Proc. Inst. Civ. Eng. Urban Des. Plan. 2009, 162, 63–70. [CrossRef] 42. Arulrajah, A.; Yaghoubi, E.; Imteaz, M.; Horpibulsuk, S. Recycled waste foundry sand as a sustainable subgrade fill and pipe-bedding construction material: Engineering and environmental evaluation. Sustain. Cities Soc. 2017, 28, 343–349. [CrossRef] Water 2020, 12, 2489 15 of 15

43. Kuriqi, A. Assessment and quantification of meteorological data for implementation of weather radar in mountainous regions. Mausam 2016, 67, 789–802. 44. Japan Meteorological Agency (JMA). Heavy Rain in July Heisei 30 and the High Temperature Characteristics and Factors Record after Mid-July. Available online: http://www.jma.go.jp/jma/press/ 1808/10c/h30goukouon20180810.html (accessed on 5 May 2019). (In Japanese) 45. Iclimate. 2018. Available online: http://www.qh323.com/portal/ClimateData (accessed on 5 May 2019). 46. Wu, H.L.; Cheng, W.C.; Shen, S.L.; Lin, M.Y.; Arulrajah, A. Variation of hydro-environment during past four decades with underground sponge city planning to control flash floods in Wuhan, China: An overview. Undergr. Space 2020, 5, 184–198. [CrossRef] 47. Xu, Y.S.; Shen, S.L.; Ren, D.J.; Wu, H.N. Analysis of factors in land subsidence in Shanghai: A view based on Strategic Environmental Assessment. Sustainability 2016, 8, 573. [CrossRef] 48. Sutanto, S.J.; Lanen, H.A.J.V.; Wetterhall, F.; Llort, X. Potential of Pan-European seasonal hydro-meteorological drought forecasts obtained from a Multi-Hazard Early Warning System. Bull. Am. Meteor. Soc. 2020, 101, E368–E393. [CrossRef] 49. Xu, Y.S.; Shen, S.L.; Lai, Y.; Zhou, A.N. Design of sponge city: Lessons learnt from an ancient drainage system in Ganzhou, China. J. Hydrol. 2018, 563, 900–908. [CrossRef] 50. Wu, Y.X.; Shen, S.L.; Yuan, D.J. Characteristics of dewatering induced drawdown curve under barrier effect of retaining wall in aquifer. J. Hydrol. 2016, 539, 554–566. [CrossRef] 51. Liu, X.; Zhou, A.; Shen, S.L.; Li, J.; Sheng, D.C. A micro-mechanical model for unsaturated soils based on DEM. Comput. Methods Appl. Mech. Eng. 2020, 368, 113183. [CrossRef] 52. Zhang, N.; Shen, S.L.; Zhou, A.N.; Xu, Y.S. Investigation on performance of neural network using quadratic relative error cost function. IEEE Access 2019, 7, 106642–106652. [CrossRef] 53. Atangana Njock, P.G.; Shen, S.L.; Zhou, A.; Lyu, H.M. Evaluation of soil liquefaction using AI technology incorporating a coupled ENN/t-SNE model. Soil Dyn. Earthq. Eng. 2020, 130, 105988. [CrossRef]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).