AOGS 2018 - Browse Abstracts

Browse by Session - Select Session:
 
Oral Presentations - Browse by Section and Presentation Day
  Section:
Poster Presentations - Browse by Section
  Section:
Input Keyword
Abstract Title Word
 

 

All Abstracts of Session HS22

Print this page

Oral Presentations

HS22 - Climate Change Risk Assessment and Adaptation on Water-related Disaster and Water Resources in Asia and the Pacific
Thursday, June 07, 2018 | 301 | 08:30-10:30
1.
HS22-D4-AM1-301-001 (HS22-A026)
 
A Review of a Climate Model Development in Japan
Akimasa SUMI#+
The University of Tokyo, Japan
#Corresponding author: sumi@ir3s.u-tokyo.ac.jp +Presenter

A program for developing a climate mode in Japan has been initiated with the Earth Simulator Project in 2002. We thought that a high-end computing power is critical for climate simulation. At the same time special attention has been paid to application to real issues. Since then, 5-year project is conducted within this program.

A historical review of our effort will be summarized and lessons will be given.

2.
HS22-D4-AM1-301-002 (HS22-A051)
 
Taiwan Climate Adaptation Technology Service (TaiCATS) – TCCIP - Team2
Ching-Pin TUNG1#+, Yung-Ming CHEN2, Jung-Hsuan TSAO1, Po Wen PERNG1
1 National Taiwan University, Taiwan, 2 National Science and Technology Center for Disaster Reduction, Taiwan
#Corresponding author: cptung@ntu.edu.tw +Presenter

Taiwan Climate Change Projection Information and Adaptation Knowledge Platform is a research project of Taiwan funded by Ministry of Science and Technology. There are three working teams involved, and each team has different goals to achieve. Team 1 of TCCIP focuses on climate modeling and analysis. The main works of Team 1 are data production, analysis and application for climate change scenarios and projection. Team 2 of TCCIP focuses on climate risk assessment and adaptation, and the main works are the development of cross-sectoral risk assessment, adaptation tools and case study. Team 3 of TCCIP focuses on climate science service integrated platform, and the main works are the integration of procedure, data, information, knowledge, wisdom and tools. In addition, Team 2 involves three working packages. The first package working on the methodology and tools of climate risk assessment and adaptation. The second package working on the test of the methodology and implementation. The third package working on demonstration case studies for land use planning considering cross sectors, multilevel governance, and cross spatial units. The goals of Team 2 are (1) to enable smart by the climate adaptation technology service and (2) to achieve future sustainable and quality life by the science support. Moreover, four deliverable outputs from Team 2 are (1) directives of climate risk assessment and adaptation plan development, (2) trans-disciplinary meta model and tools, (3) data hub with public API, and (4) climate adaptation technology service. All the outputs of Team 2 such as data, methodology, tool, and guideline will not only be used in Taiwan, but also shared with our international partners. Ultimately, Taiwan Climate Change Projection Information and Adaptation Knowledge Platform will build strong connections with international teams to develop a sustainable future together.

3.
HS22-D4-AM1-301-003 (HS22-A029)
 
Bridging Between Projection Studies and Impact Studies in Climate Change Projects
Izuru TAKAYABU1#+, Eiichi NAKAKITA2
1 Meteorological Research Institute, Japan Meteorological Agency, Japan, 2 Kyoto University, Japan
#Corresponding author: takayabu@mri-jma.go.jp +Presenter

In the TOUGOU theme C program (FY2017 – FY2021), it is requested to produce information which would contribute to the assessment of risks caused by the climate change. In the Asian region, it is indispensable to consider risks arising from heavy precipitation events and typhoons under future climate states. TOUGOU theme D uses some of these products, for concretely assessing the impact of global warming on disasters. How to produce information of climatic hazards is a main purpose of theme C program. We try the following two approaches. One is (i) to clarify the climate mechanisms and causes of the future change in the phenomena, and the other is (ii) to produce climate information applicable for impact assessment studies. For example, theme D uses very high resolution model projections around Japan Islands or South-East Asia produced in theme C, to assess various impacts, including floods, storm surges, water resources and cycles in these areas. Continuous deep communications between researchers in theme C and theme D would be needed to realize such interdisciplinary study.

This work has been conducted under the TOUGOU program of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan.

4.
HS22-D4-AM1-301-004 (HS22-A007)
 
Integrated Research Program for Advancing Climate Models (TOUGOU) - Theme D: Integrated Hazard Prediction -
Eiichi NAKAKITA1#+, Nobuhito MORI1, Kenji TANAKA1, Tetsuya TAKEMI1, Yasuto TACHIKAWA1, Toshikazu KITANO2, Hirokazu TATANO1
1 Kyoto University, Japan, 2 Nagoya Institute of Technology, Japan
#Corresponding author: nakakita@hmd.dpri.kyoto-u.ac.jp +Presenter

With the aims of further advancing climate change studies and then utilizing the results for the good of society, the Ministry of Education, Culture, Sports, Science and Technology intends to create an integrated study system with four cooperative study area themes via expansive use of the results of the Program for Risk Information on Climate Change (2012-2016), address the need for greater elucidation of climate change mechanisms, increase the sophistication of climate change pre models and assessment of the impact of climate change, and challenge ourselves to develop sophisticated climate change prediction data sets, under the name of “Integrated Research Program for Advancing Climate Models (TOUGOU)”

How will global warming affect typhoons, floods, sediment disasters, and river flows? Theme D aims to project how devastating these hazards will change over the next 100 years and scientifically reveals the relationship between global warming and disasters. Mainly the following two analysis methods will be adopted: the first one is to quantify the probability of climate change impact on typhoons and flooding etc. and the second one is to assess the impact of climate change with the worst-case scenarios that consider extraordinary situations such as super typhoons. In recent years, Japan as well as other countries have been affected by frequent and unprecedented disasters. Potential damages by such record-breaking disaters enhanced by climate change should be assessed from scientific and engineering perspectives. Moreover, we hope to provide basic information on appropriate measures needed in the future by understanding also the economic impacts.

We intend to take into account also what we need to consider and prepare for use in no-regret and proactive adaptation strategies by increasing the sophistication, integration, and depth of hazard models. In addition, we will establish the foundation for the necessary adaptation in Asian countries as well as in Japan.

5.
HS22-D4-AM1-301-005 (HS22-A052)
 
Transdisciplinary Climate Risk Assessment and Climate Adaptation Technology Service
Ching-Pin TUNG1#+, Jung-Hsuan TSAO1, Jung HUANG2, Po Wen PERNG1, Yu-Han HUANG1, Bing-Chen JHONG1
1 National Taiwan University, Taiwan, 2 National Central University, Taiwan
#Corresponding author: cptung@ntu.edu.tw +Presenter

The transdisciplinary issue of climate risk assessment and adaptation has become increasingly important. CLUS (Cross-Levels, Users and Sectors) framework in this study is an emerging methodology to cope with the complexity of climate change, and the core of CLUS is the standard risk identification procedures, which were designed base on the definition of risk by IPCC in a workable way. The authority in climate adaptation of different levels of governance can be decided by the hazard and vulnerability factors. On the other hand, the duty of adaptation can be classified specifically as central government, county government and community. Thus, the CLUS framework can be implemented in the Multi-level governance. In addition, the risk identification procedures decided by key issues of different decision-making levels and disciplines are the blueprints of trans-disciplinary meta models which aim to understand climate risks each authority needs to face and its possible adaptations. The goal of this project is to create customized products which provide different kinds of information (Data, Information, Knowledge, Wisdom: DIKW) for different users (Government, Researcher, Industry, Public: GRIP) to meet their needs. Application Programming Interface (API) and data hub for wisdom modelling play important roles in offering this kind of service. Finally, our services also include the construction and training of the meta models which could help customers to create their own user-oriented models.

6.
HS22-D4-AM1-301-006 (HS22-A012)
 
Typhoon Hazard Maps Developed by the Typhoon Ensemble Simulations
Hironori FUDEYASU1#+, Shun MIYAZAKI1, Shota YAMASAKI2, Tetsuya TAKEMI2, Masaya KATO3, Kazuhisa TSUBOKI3
1 Yokohama National University, Japan, 2 Kyoto University, Japan, 3 Nagoya University, Japan
#Corresponding author: fude@ynu.ac.jp +Presenter

This study investigates how the distribution of typhoon-related wind speeds and rainfall are enhanced by the topographical effects of their surrounding areas over the Japan Islands with the goal of creating typhoon hazard maps. To produce typhoon ensemble simulations, this study utilized the terrain shift operation method. In this method, we begin with a control run using the Weather Research and Forecasting (WRF-ARW), which is a numerical simulation performed with an initial value without the terrain shift operation. A virtual run is performed to displace the atmospheric field, defined here as the sea land/terrain distribution in the east-west (or north-south) directions at 0.2° intervals within the range of about 50° in the longitude (or latitude) direction at the initial calculation time and thereafter. As a result of sensitivity experiments to verify horizontal resolution dependence of the numerical simulations, it was confirmed that typhoon hazard maps using typhoon ensemble simulations results with horizontal resolution of 5 km can be considered to roughly reflect the influence of the terrain within areas up to several kilometers. The typhoon-related strong wind and heavy rainfalls hazard maps created using the results of the typhoon ensemble simulations of more than 800 typhoons show the distributions of average wind speed and rainfall when the typhoon center enters in the circle with a radius of 500 and 300 km at each point. For example, the hazard maps can be seen that the average wind speeds that occurred at each point associated with approaching typhoons were high in the southern areas of the Japanese Islands and weak in the northern areas. These hazard maps were possible to determine the regions where typhoon-related strong wind risks are high over the Japanese Islands, thereby providing important disaster prevention information.

7.
HS22-D4-AM1-301-007 (HS22-A017)
 
Improved Typhoon Intensity Analysis for Advanced Dvorak Technique (ADT) Using Microwave Satellite Observations.
Sungwook HONG#+, Sumin RYU
Sejong University, South Korea
#Corresponding author: sesttiya@gmail.com +Presenter

Recently, typhoon tends to gradually increase in its intensity associate with global warming. Extremely intense typhoons such as typhoon Haiyan occurred in 2013 causes socio-economic damages to many countries in Asia. In general, Advanced Dvorak Technique (ADT) using geostationary satellite observation has been widely used for analyzing the typhoon intensity because of the difficulty in direct observation of the typhoon’s internal structure. However, it is known that ADT generally underestimated typhoon’s intensity. Thus, we propose an improved ADT technique in the northwestern Pacific Ocean. Methodologically, we used the sea surface wind speed data based on SMAP (Soil Moisture Active Passive) satellite observations, and the best-track data provided by RSMC (Regional Specialized Meteorological Centre). In this study, data were collocated in both time and space with the satellite overpass in the northwestern Pacific Ocean during two years (2015-2016). The ADT with the new relationship on typhoon intensity was validated using typhoon data occurred during another two years (2013-2014). Our study showed the improved results in the analysis of typhoon intensity compared to those of the current ADT analysis. Details of methods and results will be described in oncoming presentation.

8.
HS22-D4-AM1-301-008 (HS22-A006)
 
Projections of Future Changes in Heavy Rainfall and Atmospheric Circulation Pattern in Japan During the Baiu Season by Multi-Scale Analysis
Yukari OSAKADA#+, Eiichi NAKAKITA
Kyoto University, Japan
#Corresponding author: osakada.yukari.77v@st.kyoto-u.ac.jp +Presenter

Heavy rainfall in the Baiu season is one of the main factors causing water-related disasters in Japan. As the warming trend becomes more significant, it is an urgent issue to project detailed future changes of the Baiu front-triggered heavy rainfall under climate change. Baiu heavy rainfall is a phenomenon in the small meso-β scale under atmospheric circumstances of a Baiu front in the relatively large meso-α and macro scale. Thus it is important to capture Baiu heavy rainfall from multiple spatial scales.

For multiscale analysis in our study, we used NHRCM05 and d4PDF20. NHRCM05 is a regional climate model in the high resolution of 5km, and it can provide us better heavy rainfall simulations. But, it does not have sufficient ensemble members for reliable statistical analysis. Therefore we also used d4PDF20, a regional climate model in a coarser resolution of 20km, in virtue of its huge ensemble members, consisting of 90 future and 50 present members, as well as simulation in 60 years in each member. Then, with outputs of two models, we revealed the future changes of Baiu heavy rainfall by performing multiscale analysis.  

First, we manually picked up the heavy rainfall events from NHRCM05 datasets, and obtained rainfall statistics and their future change. Meanwhile, we extracted the atmospheric patterns of pressure and vapor flux by a cluster analysis, and retrieved the occurrence frequency of these patterns from d4PDF20. The results show that the risk of Baiu heavy rainfall may increase in the Northern Japan and Japan Sea side area, where have hardly experienced Baiu heavy rainfall in the current climate, as the increasing trend of atmospheric patterns prone to heavy rainfall is proven to be significant. On the other hand, in the Pacific side area, the mechanism of Baiu heavy rainfall may also change in the future.

HS22 - Climate Change Risk Assessment and Adaptation on Water-related Disaster and Water Resources in Asia and the Pacific
Thursday, June 07, 2018 | 301 | 11:00-12:30
1.
HS22-D4-AM2-301-009 (HS22-A057)
 
Dynamical Downscaling of Typhoons Around Taiwan in Climate Projection of High-Resolution AGCM
Chao-Tzuen CHENG1#+, Hsin-Yu CHIANG1, Huang-Hsiung HSU2, Chia-Ying TU2, Akio KITOH3
1 National Science and Technology Center for Disaster Reduction, Taiwan, 2 Academia Sinica, Taiwan, 3 Japan Meteorological Business Support Center, Japan
#Corresponding author: ctcheng@ncdr.nat.gov.tw +Presenter

Climate models with resolution higher than 60km can explicitly simulate typhoons, which are one of the most concerned extreme events for Taiwan. To provide the projection of extreme events for local climate change impact studies of Taiwan, Taiwan Climate Change Projection Information and Adaptation Knowledge Platform project (TCCIP) has performed dynamical downscaling on the climate projection of high-resolution Atmospheric General Circulation Model (AGCM), the MRI-AGCM and GFDL HiRAM. Both MRI-AGCM and HiRAM project reduction in typhoon numbers for the end of 21st century in RCP8.5 scenario.  AGCMs with resolution about 50-60km can well simulate the numbers of typhoons but underestimate storm intensity too much. Although increasing the resolution to 20-30km can improve the simulation of typhoon intensity, it is too expensive to conduct ensemble simulation. In this study, dynamical downscaling using the WRF model with 5-km grids over Taiwan area is perform. No matter the driving fields are from AGCMs with 20, 25, or 50km resolution, dynamical downscaling can improved the projection of typhoon intensity, in term of winds, central pressures, and precipitation rates to reasonable values. Due to the complicated topographic of Taiwan, the added value of dynamical downscaling to local precipitation is significant, which also help to improve the impact assessment of extreme typhoon events. Success in downscaling typhoon events in ensemble simulation of 50km-meshed AGCM makes it possible to estimate the typhoon impact from a larger number of typhoon samples, resulting in more reliable statement about the future typhoon impacts due to global warming.

3.
HS22-D4-AM2-301-010 (HS22-A023)
 
Strategies on Future Climate Projections for Asian Countries and Understanding of Mechanisms of Changes in Climate Extremes in a Future Climate
Tosiyuki NAKAEGAWA1#+, Izuru TAKAYABU2, Hidetaka SASAKI1
1 Japan Meteorological Agency, Japan, 2 Meteorological Research Institute, Japan Meteorological Agency, Japan
#Corresponding author: tnakaega@mri-jma.go.jp +Presenter

We introduce our strategies on future climate projections for Japan and Asian countries under a future climate research program, Integrated Research Program for Advancing Climate Models: TOUGOU Integrated Climate Change Projection (Area Theme C). We have been developed a downscaling methodology of CMIPS's future climate projections for 10 years so that the products are useful for impact assessments. Good performances in present-day climate simulations are prerequisite for use in impact assessment but CMIPS's simulations do not meet this prerequisite. We have taken a strategies on dynamical downscaling with an atmospheric global climate model (GCM) and regional climate model (RCM). This strategy allow us to perform the climate simulations with high reproducibility of present-day climate as well as with high horizontal resolutions. This strategy does not only meet the prerequisite but also provide opportunities in understanding mechanism of changes in climate extremes in a future climate because atmospheric phenomena on a fine scale are well reproduced. For example, heavy rainfalls such as annual maximum daily precipitation and 3-day precipitation total at the late 21st century are projected to increase in most Asian countries with robustness information; frequency of total typhoons in the future climate are projected to decrease due to stable atmospheric conditions in the troposphere and the frequency of strong typhoons are projected to increase. This strategy has been applied to the future climate projections in Japan for the decade, but now have been applied to those of Asian countries as research collaborations under a dynamical downscaling program, COordinated Regional Downscaling Experiment (CORDEX) East Asia and Southeast Asia. A total of 23 researchers from 7 countries have been invited for 6 years. They conducted future climate projections over their own countries with high horizontal resolutions by using a Non-Hydro RCM (NHRCM) for climate projections and impact assessments.

4.
HS22-D4-AM2-301-011 (HS22-A025)
 
River Discharge Simulation by a Distributed Hydrologic Model Utilizing NHRCM 5km Output in Thailand
Aulia Febianda ANWAR TINUMBANG1#+, Kazuaki YOROZU1, Yasuto TACHIKAWA1, Yutaka ICHIKAWA1, Hidetaka SASAKI2
1 Kyoto University, Japan, 2 Japan Meteorological Agency, Japan
#Corresponding author: anwar.febianda.28n@st.kyoto-u.ac.jp +Presenter

The impact of climate change under global warming is clearly happening: the average global temperature is increasing, the sea level is rising, the numbers of extreme rainfall events have increased, etc. Therefore, it is important to make a risk assessment for the current and future climate change. To conduct hazard assessment, it is necessary to have a detailed projection of climate data. Now, a result of first simulation trial of NHRCM 5km-resolution is available for Thailand area, provided by Meteorological Research Institute of Japan Meteorological Agency (MRI/JMA). The aim of this research is to evaluate the inflow river discharge in Bhumibol dam, by comparing the simulated river discharge using NHRCM data with the observed data. Before the river discharge evaluation is performed, the 20-years-average rainfall data of NHRCM was compared with APRHODITE data. We found that NHRCM rainfall value in the most of region was underestimated. However, the northern region, which almost corresponds to Bhumibol dam catchment, was overestimated, particularly in the beginning of rainy season. Hence, NHRCM rainfall value shows better accuracy. After that, we calculated runoff value using NHRCM data by utilizing land surface model, called SiBUC model. Then, the output of runoff was used as an input to calculate river discharge using flow routing model (1K-FRM). We compared the 20-years-average simulated river discharge and the observed value by drawing a flow-duration curve (FDC), with a daily time interval. It is found that during normal, low, and drought-flow period, which is equaled to 2nd, 3rd, and 4th quartile of FDC, respectively, the simulated flow was well compared with the observed data. Only in the first quartile, or high-flow period, the river discharge was about 15% overestimated by NHRCM. Overall, the simulated river discharge is consistent with the result of rainfall analysis.

5.
HS22-D4-AM2-301-012 (HS22-A003)
 
Quantitative Assessment of Climate Change Impacts on Flood Risk in Davao Oriental, Philippines
Jonathan CABRERA1,2#+, Han Soo LEE1
1 Hiroshima University, Japan, 2 Davao Oriental State College of Science and Technology, Philippines
#Corresponding author: jonathan.cabrera@doscst.edu.ph +Presenter

This study aims to quantitatively assess the climate change impacts on flood-prone risk areas in Davao Oriental, Philippines, for the future in 2030, 2050, and 2100 in comparison with the present situation by identifying flood risk zones based on the multi-source data, including rainfall, slope, elevation, drainage density, soil type, distance to the main channel and population density. The temperature and rainfall projections for the future are utilized from the Coupled Model Inter-comparison Project Phase 5 (CMIP5) predictions of IPCC AR5. The future temperature from the CMIP5 predictions showed that Davao Oriental experiences approximately 1℃ and 3℃ increases under RCP 4.5, and RCP 8.5 scenarios, respectively. However, the rainfall slightly increases in the coming years. Among the thirty-nine GCM models available from CMIP5, the GFDL-ESM2M model showed a good agreement to the observed rainfall dataset at local stations. The intensity of rainfall increases at approximately 69% in the future, resulting in the increase of the magnitude of flooding. The flood risk map shows that 95.91% of Davao Oriental is under low and moderate flood risk categories for the present and those categories slightly decrease to 95.75% in the future. The high and very high flood risk areas cover approximately 3% of the province for the present and show no dramatic change in the future. Twenty-three out of the one hundred eighty-three barangays (towns) are at high and very high risks of flooding for the present whereas, in the coming years, only one barangay is in very high risk of flooding. These barangays under the high and the very high risks of flooding are primarily situated on riversides and coastal areas. Thus, this needs the immediate actions of the decision-makers to develop a community-based disaster risk plan.

6.
HS22-D4-AM2-301-013 (HS22-A036)
 
Future Flood Simulation in Midlatitude Region (Hokkaido) Using High-Resolution Heavy-Rainfall Data
Nobuaki KIMURA#+, Hirohide KIRI
National Agriculture and Food Research Organization, Japan
#Corresponding author: kimuran590@affrc.go.jp +Presenter

Severe flood that occurred in Eastern Hokkaido (Japan) in August 2016 because of continuous hitting of plural typhoons caused catastrophic damages over upland fields. A similar flood-induced disaster may often occur in the future climate change, affected by global warming. Such a potential disaster in the future raises serious concern in the midlatitude region, where has seldom experienced severe flood. It is necessary to understand a severity and scale of the future flood to reduce disaster risks and implement appropriate adaptations. In this study, we have conducted flood simulations at the occurrence of the flood disaster in Eastern Hokkaido in August 2016 (hereinafter, 2016 East-Hokkaido flood disaster) for the calibration of a hydrological model. The hydrological model employs conceptual, distributed rainfall-runoff analysis module and a GIS data-input function. The information of future flood simulation is also provided using the heavy rainfall data generated by the future projection (2075-2099) of a high-resolution three-dimensional atmosphere-ocean coupled regional model, considering a similar atmospheric environment of the 2016 East-Hokkaido flood disaster. A compassion between current and future floods reveals how much the severity and the scale of the disasters are enhanced over the midlatitude agricultural region in the future. In addition, the agricultural loss over the damaged upland fields can be estimated, based on the ratio of future to current food severity and scales.

7.
(HS22-A021)
 
High-Res. Climate Simulation and Projection of HiRAM-AGCM with Focus on Extreme Weather
Yu-Shiang TUNG1#, Huang-Hsiung HSU2+, Chao-Tzuen CHENG1, Akio KITOH3, Chia-Ying TU2
1 National Science and Technology Center for Disaster Reduction, Taiwan, 2 Academia Sinica, Taiwan, 3 Japan Meteorological Agency, Japan
#Corresponding author: yushiangtom@gmail.com +Presenter

Taiwan features a complex topography characterized by mountainous regions and rugged terrain, where rivers are short, valleys are narrow and geological features are fragile.  It is also frequently attacked by western North Pacific (WNP) typhoons. Thus, high resolution climate simulation is needed to accurately assess extreme weather variations and changes. Taiwan Climate Change projection and adaptation Information Platform (TCCIP) project has been using outputs of MRI-AGCM and GFDL HiRAM for driving WRF in dynamic downscaling simulations (5 KM resolution) to resolve the effect of complex terrain on extreme weather and to obtain more reliable results for Taiwan regional climate change assessment and application. Both AGCM simulations are forced using same sea surface temperature variation and changes under RCP8.5 scenario. A comparison of two sets of result yield more confidence to the projected future change in extreme weather events. Dynamic downscaling approach is found to be able to effectively correct the biases produced by AGCM with 20-25 KM resolution and therefore yield more reliable projection results. Overall speaking, the extreme events will become stronger in amplitude. More results will be reported in the presentation.

HS22 - Climate Change Risk Assessment and Adaptation on Water-related Disaster and Water Resources in Asia and the Pacific
Thursday, June 07, 2018 | 301 | 13:30-15:30
1.
HS22-D4-PM1-301-014 (HS22-A031)
 
Future Change Analysis of Extreme Floods Using Large Ensemble Climate Simulation Data
Yasuto TACHIKAWA1#+, Tomohiro TANAKA1, Kohei MIYAWAKI1, Kazuaki YOROZU1, Yutaka ICHIKAWA1, Sunmin KIM2, Masaya KATO3
1 Kyoto University, Japan, 2 -, Japan, 3 Nagoya University, Japan
#Corresponding author: tachikawa@hywr.kuciv.kyoto-u.ac.jp +Presenter

Future changes of probability distributions of extreme rainfall and flood are analyzed using “Database for Policy Decision-Making for Future Climate Change, d4PDF”' having large ensemble members of climate simulations for the 60-years historical simulations with 50 emsemble members and 60-years future simulations with 90 emsemble members. The magnitude of a largest-class floods equivalent to a 1000-years flood is also analyzed. The study basins are the Ara River basin (2940km2) in the Tokyo metoroporitann area, the Shonai River basin (1010km2) in the Nagoya area, and the Yodo River basin (8240km2) in the Osaka and Kyoto area in Japan. The results show that 1) the frequency distributions of annual maximum 24-hours rainfall for d4PDF of the historical experiments match with the ones for observed data; 2) the 200-years annual maximum 24-hours rainfall for the d4PDF of the future experiment is 1.3 to 1.4 times larger than the one of the historical experiment and 1.5 to 1.7 times larger for the annual maximum river discharge; 3) the 200-years annual maximum 24 hours rainfall for the d4PDF of the future experiment is equivalent to the one of 900 years of the historical experiment; and 4) the rainfall patterns that cause largest-class floods simulated by d4PDF well match with the one of large historic floods for the three basins.

2.
HS22-D4-PM1-301-015 (HS22-A020)
 
Changes in Future Flood Estimation Under Climate Change Scenarios in Han-River Basin, South Korea
Sunghun KIM+, Younghun JUNG, Hyunjun AHN, Jun-Haeng HEO#
Yonsei University, South Korea
#Corresponding author: jhheo@yonsei.ac.kr +Presenter

Extreme flood disaster is one of the most important issue on the climate change impact. Increasing quantity of precipitation, and associated change of rainfall frequency represent a serious hazard to the infrastructures in Korea, and all over the world. Intergovernmental Panel on Climate Change (IPCC) provided the new Representative Concentration Pathway (RCP) scenarios in 2014. A lot of studies have been performed with RCP scenarios to analyze the influence of climate change on the hydrologic process. According to previous researches, the annual average precipitation and number of days of a heavy rainfall are expected to increase in Korea. However, the actual hydrologic structures are not prepared against the climate change systematically, and the quantitative analysis is essential to prevent the future flood risk. Consequently, we compared the estimated current flood with projected future flood. The current flood is estimated by using observed rainfall data, and the future flood is estimated by using precipitation data from climate models. The regional climate models (RCMs) from the Coordinated Regional Downscaling Experiment in East Asia (CORDEX-EA) were used for future climate change projection in Korea. The RCMs simulated both the historical climate (1979-2005) and future climate projection (2006-2050) under the RCP4.5 and RCP 8.5 scenarios. In this study, Han-river basin is selected as the subject area, a representative watershed in Korea. The flood quantile is estimated using HEC-1 model. The results of flood quantiles by several RCMs are compared and analyzed in this study.

3.
HS22-D4-PM1-301-016 (HS22-A038)
 
Impact of Climate and Land Cover Changes on Flooding in a Humid Tropic River Basin in Sumatra, Indonesia
Takahiro SAYAMA1#+, Kodai YAMAMOTO1, Apip APIP2, Kaoru TAKARA1
1 Kyoto University, Japan, 2 Indonesian Institute of Sciences, Indonesia
#Corresponding author: sayama.takahiro.3u@kyoto-u.ac.jp +Presenter

In Sumatra in Indonesia, large-scale plantations of oil palm and acacia trees have caused 50 % reduction of natural forests in the 25 years between 1985 to 2009. Both climate and land cover changes may impact on regional water cycle, which potentially leads to increase the risk of flood and drought disasters. Hydrologic process understandings and their representations by numerical models are therefore important scientific steps for achieving adequate assessment of climate and land cover changes. Among various hydrologic processes, rainfall-runoff processes in humid tropics typically characterized by deep weathered soil layers have been still poorly understood. Based on field investigations and modeling, this study first attempts to simulate rainfall-runoff and flood inundation processes at the river basin scale in the Batanghari River basin in Sumatra, Indonesia. In terms of the field investigation, we monitor hillslope groundwater dynamics in the central part of the basin covered by tropical forest. The latest monitoring data suggested rapid response of groundwater even in the deep soil layer. By reflecting such conditions in the model, it conducts a long-term hydrologic simulation with climate change projections including MRI-AGCM outputs to estimate the high potential regions of flooding under climate and land cover changes in this region.

4.
HS22-D4-PM1-301-017 (HS22-A047)
 
On Consecutive-Storm Event Based (ConSEB) Model for Short Term Flood Runoff Simulation
Duck Hwan KIM, Hung Soo KIM#+
Inha University, South Korea
#Corresponding author: sookim@inha.ac.kr +Presenter

Sometimes, two or three strong storm events or extreme events were continuously passing through a certain region or rainfall storm had a long term duration such as 10 or 20 days and the severe flood damages due to those events were occurred. The purpose of this study is to develop a model which can analyze consecutive rainfall events or an event with long term duration. Mostly we have been used HEC-HMS model for flood runoff simulation with a storm rainfall event and so we tried to simulate flood runoff for two or more storm events which were continuously occurred using HEC-HMS model. Then we figured out the limitations of HEC-HMS model for the runoff simulation with two or more events and developed the hourly direct runoff estimation method based on the runoff curve number(CN) which can be changed in climate condition to overcome the limitations of HEC-HMS model. Furthermore, the continuous rainfall analysis is possible in conjunction with the base flow estimated by the Eckhardt Filter. The study area is Gyeongan stream basin, Korea. Applicability of the developed model was carried out using the rainfall event in july, 2009 among the maximum rainfall events in Gyeongancheon stream basin. The difference between the peak runoffs was 34.5 CMS, the percent error in peak was 0.9925, RMSE was 165.1071, and Nash-Sutcliffe efficiency was 0.9371. The results seemed to be a quite excellent and so the developed model in this study can be used for consecutively occurred storm events or events having a long term duration.

Keywords : Consecutive storm events, CN, Eckhardt filter, ConSEB model

Acknowledgements

This work is supported by the Korea Agency for Infrastructure Technology Advancement(KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant 18AWMP-B083066-05).

Corresponding author information: Hung Soo Kim, Department of Civil Engineering, Inha University, Yonghyun-Dong, Nam-gu, Incheon, 402-751, South Korea. E-mail: sookim@inha.ac.kr; Phone:+82-32-860-7572; Fax:+82-32-876-9787.

5.
HS22-D4-PM1-301-018 (HS22-A054)
 
Analysis of the Return Period of Flash Flood in Small Mountainous Basins Under Climate Change
Hwa Yeon KIM+, Deg-Hyo BAE#
Sejong University, South Korea
#Corresponding author: dhbae@sejong.ac.kr +Presenter

The mountainous headwater basins in South Korea have a high risk for flash flood occurrence due to the rapid increase of water levels at the basin outlet caused by the geographical characteristics of narrow channel width and steep sloping terrain. It is therefore important for estimating the vulnerability of flash flood occurrence in order to reduce the flood damages of people and property. There are several approaches for developing flash flood forecasting model. One of the typical methods is the use of flash flood guidance (FFG), which is composed of the threshold runoff model and soil moisture estimation model. On the other hand, we expect that the future climate change will affect the occurrence of flash flood. The aim of this study is to provide a method to analyze the climate change impact on flash flood occurrence. The study area is the 200 small mountainous basins of the Han River in South Korea. The river basin and channel characteristics are extracted for the computation of bankful flows and the threshold runoffs for each sub-basin are computed based on the Manning formula and geomorphologic unit hydrograph. And then, the frequency analysis is performed to find out the corresponding return period of precipitation for occurring bankful flow at each drainage basin outlet. For the evaluation of the change of return period occurring flash flood according to climate change, CMIP5 data and statistical and dynamic downscaling schemes are used in this study. We will deliver how the climate change will affect the change of return period for the flash flood occurrence.

Acknowledgement

This work is supported by the Korea Agency for Infrastructure Technology Advancement(KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant 18AWMP-B127555-02).

6.
HS22-D4-PM1-301-019 (HS22-A011)
 
Flood Risk Assessment and Adaptation Under Extreme Climate Scenarios in Tainan City, Taiwan
Yi-Chiung CHAO1#+, Yi-Hua HSIAO1, Lun-Tsun CHEN2, Chih-Tsung HSU2, Keh-Chia YEH3, Chao-Tzuen CHENG1, Hsinchi LI1
1 National Science and Technology Center for Disaster Reduction, Taiwan, 2 National Center for High-performance Computing, Taiwan, 3 National Chiao Tung University, Taiwan
#Corresponding author: ycchao@ncdr.nat.gov.tw +Presenter

Flood disaster has been increased in last decade, and has usually caused huge economic loss in Taiwan, especially in Tainan City. The Tainan City government recently enhanced its overall flood protection level. These mitigation measures were expected to reduce flooding damage; however, the reducing effects might be limited under climate change. Therefore, this study aims at understanding flood risk on current situation under global warming scenario, and proposes adaptation methods to reduce risks in Tainan City.

Typhoons simulated by the very high-resolution AGCM of Meteorological Research Institute (MRI-AGCM3.2S) were dynamically downscaled to 5-km resolution to represent the projection of extreme rainfall events of the present day (197-2005) and the end of 21st century (2075-2099). Top 30 typhoons based on their regional-averaged total rainfall amount were selected for each period. SOBEK, a two dimensional flood simulation model, was used to assess the flood risk. Increasing levee protection level and pumps were used as adaptation measures to understand the effectiveness of flood damage countermeasures in the end of 21st century.

The results show that 24-hour maximum cumulative rainfall of 30% (90%) typhoon events of the present day (the end of 21st century) will exceed drainage system protection level. Moreover, mean inundation area induced by extreme rainfalls is three times more in the end of 21st century than in present day. It implies the current mitigation methods will not be sufficient under extreme rainfall events. Unfortunately, the adaptation applied in this study for flood hazard in the end of 21st century was also inefficient; since the average inundation area decreased only 2.6% in the most extreme typhoon event. It implies that the traditional adaptation measures are limited when we face flood disaster in the future; more innovation to enhance adaptation will be needed. Meanwhile, improving local disaster resilience and risk awareness are necessary.

7.
HS22-D4-PM1-301-020 (HS22-A024)
 
Landslide Risk Assessment for Various Land-Use Categories Under Climate Change
Chi-Wen CHEN+, Tingyeh WU#
National Science and Technology Center for Disaster Reduction, Taiwan
#Corresponding author: kuramaimissyou0@gmail.com +Presenter

This study aims to discuss the risk assessment and adaptation strategies of climate change impacts on landslide disasters. Many studies has adopted risk assessment models and established risk maps. However, these results are difficultly applied for adaptation owing to the unclear definition of landslide risk. The connection between risk assessment and adaptation is not only the most important part, but also the main purpose when discussing climate change impacts. Therefore, this study focuses on the connection between adaptation process and risk assessment, clarifying the definition of each factor in the risk assessment and separating into two analysis stages. The first stage is to identify regions with high landslide impacts under climate change and the second stage is to discuss the detailed risk assessment and adaptation strategies in those regions. Here we present results at the first stage.  

During the first stage, landslide hazard-exposure maps under climate change with different land-use categories are generated. The hazard is analyzed using effective rainfall variations from the end of 20th century to 21th century and landslide hot spot values. The effective rainfall variation is the difference of the 99.5 percentile values of hourly effective rainfall between the two periods, and the landslide hot spot is calculated by the landslide frequency during 2005 and 2015. The result shows that the landslide hazard under climate change is higher in central and southern areas of Taiwan than those in northern and eastern areas. The distribution of landslide hazard under climate change in Taiwan is used as a basic map to overlay with the distribution of different land-use categories and produce landslide hazard-exposure maps. By this process, regions with high landslide hazard-exposure in many land-use types are identified, where will be chosen as sample areas for the second stage with detailed discussion on risk assessment and adaptation strategies.

8.
HS22-D4-PM1-301-021 (HS22-A040)
 
Vulnerability and Adaptation of Glacier Change in China
Jianping YANG#+
Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, China
#Corresponding author: jianping@lzb.ac.cn +Presenter

The arid region of Northwest China has a unique glacier-oasis landscape which is glaciers in mountainous area and oasis in the middle and lower reaches. Glacier melt water is an important supply for socio-economic development of oasis. Thus, changing glaciers is going to significantly influence sustainability of oasis system in the future. Hexi inland river basin is selected as an example for studying the vulnerability and adaptation of oasis social-ecological system affected by glacier change in the arid region of Northwest China.Hexi inland river basin is composed by three basins from Shiyanghe river in the east through Heihe river in the middle to Shulehe river in the west. The vulnerability of the oasis social-ecological system to glacier change has an increasing tendency in the study area in a basin scale during the 1995 to 2010. Among three basins, the Shiyang river basin is the most affected by glacier changes, and the vulnerability is the highest, followed by Heihe river basin, and finally the Shule river basin. On a county scale, more than moderate vulnerable areas account for 47.62% of the total counties in Hexi inland river basin. These vulnerable areas are mainly distributed in the oasis area in the middle of the basin. Rapid developments of society and economy significantly increase exposure of the oasis system to glacier change. That is the major reason of high vulnerability in the oasis area. To adapt glacier change the oasis system should enforce the following initiatives and strategies: (1) delineating the red line of oasis area, (2) controlling population, (3) implementing high efficiency water saving, (4) adjusting the traditional mode of agriculture leading oasis economy, heightening the development and utilization of glacier service function in upstream, and promoting ecological service function in oasis area.

HS22 - Climate Change Risk Assessment and Adaptation on Water-related Disaster and Water Resources in Asia and the Pacific
Thursday, June 07, 2018 | 301 | 16:00-18:00
1.
HS22-D4-PM2-301-022 (HS22-A008)
 
Seamless Climate Change Impact Assessment Until the End of 21st Century
Kenji TANAKA#+
Kyoto University, Japan
#Corresponding author: tanaka.kenji.6u@kyoto-u.ac.jp +Presenter

In this presentation, research framework and targets of the "Seamless Climate Change Impact Assessment Until the end of 21st Century" group is briefly introduced. This research group consists of 7 subgroups ; (1) continuous hydrological cycle change assessment, (2) advanced reservoir operation under climate change, (3) water related impact on agricultural productivity, (4) water quality change assessment, (5) continuous coastal hazard assessment, (6) continuous coastal environment change assessment, (7) coupling GCM and land surface model with inclusion of human activity.

The main future climate information for this group is 150 years continuous simulation of NHRCM20 down-scaled from MRI-AGCM3.2H (AGCM60). As the 150 years continuous outputs are not available so far, impact assessment models are developed/improved/calibrated using the outputs from previous project "SOUSEI".

Some of the preliminary research activities and results will be introduced.

2.
HS22-D4-PM2-301-023 (HS22-A009)
 
Water Resources Risk Assessment of Northern Taiwan for Climate Change Adaptation
Tzu-Ming LIU1#, Ming-Hsu LI2+, Ching-Pin TUNG3
1 National Science and Technology Center, Taiwan, 2 National Central University, Taiwan, 3 National Taiwan University, Taiwan
#Corresponding author: tedliu13@gmail.com +Presenter

Climate change is a phenomenon we cannot ignores its effects and has become increasingly evident worldwide. In Taiwan, climate change is a pressing fact and a lot of people have suffered the consequence of it in their daily life. Water resources is the one directly under the threat of climate change. To against climate change, it is very important to find where the most vulnerable area is and adapt the whole water resources system from it. To have robust and resilient water resources system, this study will use a systematic method to identify the key issue of water resources to climate change, evaluate the current and future risk, and identify the key risk of water resources. To simulate the impact of climate change to water supply system of northern Taiwan, the hydrological model within GWLF was used to evaluate the hydrological impact of four watersheds, Feitsui Reservoir watershed (Beishi river), Nanshi river watershed, Shihmen reservoir watershed (Dahan river) and Sanxia river watershed. The simulated discharge of four watersheds were put into the northern Taiwan water resources system dynamics model to simulate the water supply and demand in the northern Taiwan areas (Taipei, Banxin and Taoyuan) and explore the risks of water resources. The key risk will be indicated, and the adaptive strategies will be applied to strengthen the water resources system.

3.
HS22-D4-PM2-301-024 (HS22-A013)
 
Evaluation of Water Supply Capability for Drought Risk Management Considering Reserve Storage on Multi-Purpose Dam
Jinhyeog PARK1#+, Suhyung JANG1, Shinuk KANG1, Youngteck HUR1, Jungmin KIM1, Hyeongung KANG1, Jaeeung YI2
1 K-water Institute, South Korea, 2 Ajou University, South Korea
#Corresponding author: park5103@kwater.or.kr +Presenter

The average frequency of significant droughts in Korea was about 5 to 7 years, but it has found almost every year since 2006. The multi-purpose dams, which are managed and operated by K-water, cover more than 60% of domestic water supply. The water supply from multi-purpose dams during the dry season has been too much dependant on the inflow volume during the flood season. In other words, the multi-purpose dams have to make a room for flood control during flood season while water supply for dry season totally depends on the water stored during flood season. However, the spatio-temporal variability of precipitation is significantly diverse over the regions and seasons. Also, the water shortage would be accelerated caused by climate change. Each multi-purpose dam has an emergency storage which can be used in case of an urgent water shortage condition, and can support the sustainable water supply effectively. Up to now it has not been being utilized because there is no guidance or manual in order to incorporate into the drought mitigation and response plans. In this study, the concept and methodology of hedging rules with water supply reserve on multi-purpose dams are introduced as a non-structural measure for drought risk management. The multi-purpose dams in Nakdong River Basin, where is relatively vulnerable to drought risks in Korea, was selected to evaluate the performance of water supply capability based on the inflow conditions from AR5 GCM(CMCC-CMS) RCP 4.5 scenario. To this end, the optimal water supply reserve is first estimated and then integrated into the revised reservoir operation rules for each multi-purpose dam. As a result of this study, it is found that the multi-purpose dams could preserve the sustainable water supply under future climate change conditions and it is expected that the revised reservoir operation rules with water supply reserve concept could support more effectively to satisfy the water demands and to minimize the water shortage conditions.

Acknowledgements

This work is supported by the Korea Agency for Infrastructure Technology Advancement (KAIA) grant funded by Ministry of Land, Infrastructure and Transport (Grant No. 18AWMP-B083066-05).

4.
HS22-D4-PM2-301-025 (HS22-A027)
 
Comparative Standardized Precipitation Evapotranspiration Index Analysis of d4PDF_GCM Dataset
Akira HASEGAWA1,2#+, Maksym GUSYEV1
1 Public Works Research Institute, Japan, 2 the University of Tokyo, Japan
#Corresponding author: hase55@pwri.go.jp +Presenter

We developed the comparative standardized precipitation index (cSPI) computed using future climate precipitation datasets on the basis of the present climate precipitation to evaluate meteorological drought under climate change (Hasegawa et al., 2016). The cSPI is computed only with precipitation dataset using standardized index procedure of SPI and has been applied on a river basin as well as global scales. Despite its usefulness, it is not designed to account for potential or actual evapotranspiration changes. Therefore, we expanded the cSPI approach to the comparative standardized precipitation evapotranspiration index (cSPEI), which is computed as a difference between precipitation and evapotranspiration, in this study.

The global part of “database for Policy Decision making for Future climate change” (d4PDF_GCM) consists of outputs from global warming simulations by a global atmospheric model with horizontal grid spacing of 60 km. The d4PDF_GCM includes three 60-year sets of climate experiments: historical and non-warming climates (1951-2010) with 100 ensemble members each, and +4K climates warmer than the pre-industrial level (2051-2110) with 90 members. Computing both cSPI and cSPEI with the d4PDF_GCM dataset, we can investigate the difference between the cSPI and cSPEI and evaluate the drought changes under +4K and non-warming climates on the basis of the historical climates with considering uncertainty due to differences of initial conditions and sea surface temperature patterns (+4 K only).

References

Hasegawa A., Gusyev M.A., and Y. Iwami (2016). Meteorological Drought and Flood Assessment using the Comparative SPI Approach in Asia under Climate Change. Journal of Disaster Research 11(6): 1082-1090, doi: 10.20965/jdr.2016.p1082.

5.
HS22-D4-PM2-301-026 (HS22-A030)
 
Application of Climate Scenario Planning to Assess Vulnerability of Water Supply in Nakdong River Basin, Korea
Si-Jung CHOI#+, Seongkyu KANG, Dong-Ryul LEE
Korea Institute of Civil Engineering and Building Technology, South Korea
#Corresponding author: sjchoi@kict.re.kr +Presenter

Recently, there are many cases where it is difficult to supply stable water due to severe drought in Korea. It is expected that stable supply of water will be difficult in the future due to climate change. It is necessary to enhance the responsiveness of water supply through the impact assessment. However, it is difficult to accurately predict the future situation with large uncertainty, so it is hesitant to take measures. There are not many studies quantitatively evaluating the impact of climate change on water supply, so there is a limit to providing related information. The water resources plan prepares and analyzes the problems that may arise in the future rather than the present, and it can be done by analyzing the scenarios that can take into consideration various situations in the future. The purpose of this study is to construct various future scenarios reflecting climate change and social/economic condition change such as population change, economic growth rate that can occur in future. We evaluated the impact of climate change on water supply capacity of the water resources facilities and the vulnerability of the water supply network located in the Nakdong River basin of Korea. As a result of analysis, difficulties in supply of water due to climate change are anticipated in the Nakdong River basin of Korea, it is judged that various countermeasures are needed to solve this problem. 

Acknowledgement

This work is supported by the Korea Agency for Infrastructure Technology Advancement(KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant 18AWMP-B083066-05).

6.
HS22-D4-PM2-301-027 (HS22-A033)
 
Impact of Climate Change on Rice Production and Strategies for Adaptation in Taiwan
Ming-Hwi YAO#+
Taiwan Agricultural Research Institute, Taiwan
#Corresponding author: mhyao@tari.gov.tw +Presenter

Taiwan's main crop is rice, and any change in rice yield is critical for food security. In the face of future climate change, accurately estimating the yield of rice and performing related adjustments is crucial. Assessing crop yield and climate factors typically rely on crop modeling, and the DSSAT software is widely used in rice-producing countries for assessing the impact of climate change. In this study, a food production assessment system was developed, with DSSAT used to integrate a database containing rice growth parameters relevant to Taiwan. By analyzing future climate conditions predicted using RCP2.6, RCP4.5, RCP6.0 and RCP8.5 from IPCC AR5, the simulation results were then input into a geographic information and systematic function to divide Taiwan into 1568 grid points according to the estimates of 5 × 5-km grids, and the differences in rice production yield among various grid locations were assessed. Under the scenarios of RCP2.6, RCP4.5, RCP6.0 and RCP8.5, the yield also shows a decreasing trend toward the end of the century, an average reduction in rice yield by 6%, 12%, 14% and 23% in the first crop harvest season, and by 7%, 14% 18% and 26% in the second harvest season, respectively. There are generally two harvest seasons in Taiwan. The first harvest season is from February to June, during which time the average temperature is lower, the number of days to growth is 120, and the yield is higher compared with the second harvest season (August–November). An effective adjustment strategy could assist in adjusting the seeding date according to the future climate conditions and also could avoid the heavy rain which usually occurs in June. The result shows advancing the planting date for the first crop harvest season by 30 days could increase the yield by 438 kg ha-1.

7.
HS22-D4-PM2-301-028 (HS22-A045)
 
Progressive Assessment of Future Climate and Land Use Changes Impact on Watershed Hydrology and Stream Water Quality Using SWAT
Ji-Wan LEE#+, Seong-Joon KIM, Chung Gil JUNG
Konkuk University, South Korea
#Corresponding author: closer01@konkuk.ac.kr +Presenter

This study tries to evaluate the future climate and land use changes impact on watershed hydrology and stream water quality of Geum river basin (9,645.5 km2) in South Korea with decadal accumulation changes from baseline such as 2010-2019, 2010-2029, 2010-2039, and so on. The SWAT was set by dividing the whole basin into 78 sub-basins including 2 multipurpose dams (YDD and DCD) and 3 big operation weirs (SJW, GJW, and BJW) within the watershed. The SWAT was spatially calibrated and validated using the 2005~2015 observed data of daily dam inflows and weir water levels, flux tower evapotranspiration (ET), Time Domain Reflectometry (TDR) soil moisture contents (SWC), groundwater levels (GWL), and 8 days interval of suspended solid (SS), total phosphorus (T-P), and total nitrogen (T-N) stream water quality. The average Nash-Sutcliffe efficiency (NSE) for dam inflows was 0.64. The average determination coefficient (R2) for evapotranspiration, soil moisture content, and ground water level was 0.52, 0.60, and 0.59 respectively. The average R2 of SS, T-P, and T-N was 0.77, 0.70, and 0.63 respectively. Based on the watershed hydrology and stream water quality of 1975~2015 baseline period, the future changes were evaluated using RCP (Representative Concentration Pathway) 8.5 climate change scenarios of HadGEM2-ES, INM-CM4, FGOALS-s2, and HadGEM3-RA and CLUE-s (Conservation of Land Use and its Effects at Small regional extent) land use change scenario. By the decadal accumulation evaluation, we could trace the progressive changes of future hydrology and water quality, and the results showed that the future watershed hydrology are mainly affected by the climate and the future stream water quality are largely affected by the human activity. The future urban-growing land use change under the future climate change showed considerable impact on watershed hydrology and the consequent stream water quality.

Acknowledgements

This work is supported by the Korea Agency for Infrastructure Technology Advancement (KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant 18AWMP-B083066-05).

8.
HS22-D4-PM2-301-029 (HS22-A049)
 
Improved Confidence in Drought Projections over Korea Based on the Multiple Climate Change Scenarios and Multiple Drought Indices
Moon-Hwan LEE1+, Eun-Soon IM1#, Deg-Hyo BAE2
1 The Hong Kong University of Science and Technology, Hong Kong SAR, 2 Sejong University, South Korea
#Corresponding author: ceim@ust.hk +Presenter

Given that climate change is expected to intensify the hydrological cycle with greater variability, many efforts have been made to predict where and how severe the droughts might occur over Korea. However, the projected intensity and the regional extent of future drought are still a subject of debate due to a large uncertainty. Long-term projection of drought under ongoing global warming seems to be considerably dependent on the selection of climate models and emission scenarios. Furthermore, an absence of any universally superior drought index imposes the additional uncertainty. In this study, we assess the future changes in drought characteristics for four major river basins in Korea based on multiple climate change scenarios and multiple drought indices. To better resolve the region-specific climate changes and to enhance the confidence of future projections, two global climate model projections (HadGEM2-AO and ACCESS1.0) forced by two Representative Concentration Pathway (RCP4.5 and RCP8.5) scenarios are dynamically downscaled using two regional climate models (RegCM4 and WRF). While dynamically downscaled climate projections are directly used for the calculation of Palmer drought severity index (PDSI) and standardized precipitation index (SPI), they are also provided for the input to run hydrological model (VIC) for generating the runoff which is used for the calculation of standardized runoff index (SRI). Comparing these drought indices can provide an opportunity to comprehensively understand the behaviors of meteorological and hydrological droughts in response to global warming, and to build on previous findings that assessed similar topic. 

Acknowledgements

This research is supported by the Korea Agency for Infrastructure Technology Advancement (KAIA) grant funded by the Ministry of Land, Infrastructure, and Transport (Grant 18AWMP-B083066-05). WRF and RegCM4 downscaled projections driven by HadGEM2-AO were provided by Pusan National University (Prof. Joong Bae Ahn) and Kongju National University (Prof. Myoung-Seok Suh), respectively.

HS22 - Climate Change Risk Assessment and Adaptation on Water-related Disaster and Water Resources in Asia and the Pacific
Friday, June 08, 2018 | 301 | 08:30-10:30
2.
HS22-D5-AM1-301-030 (HS22-A056)
 
Re-Evaluation of Boryeong Dam Conduit Project’s Economic Feasibility with Real Options Analysis
Seung Beom SEO1#+, Sun Hoo IHM2, Young-Oh KIM1
1 Seoul National University, South Korea, 2 Korea Environment Institute, South Korea
#Corresponding author: sbseo7@snu.ac.kr +Presenter

To overcome the impact of the persistent drought that occurred recently in Korea, the province of Chungcheongnam-do constructed the conduit from the lower stream of Geum river Baekjae-bo to the upper stream of Boryeong Dam in 2016. However, due to rising uncertainties in climate change, not only was it difficult for the government to decide whether to proceed with the project, but a conflict also arose due to the local public’s opposing opinion regarding the project’s feasibility. Therefore, in this study, we re-evaluated the economic feasibility of the project by utilizing a real options analysis (ROA) that can consider the uncertainties in climate change. First, we evaluated the economic feasibility of the project under climate change uncertainty through the application of a traditional benefit-cost analysis method. Through the consideration of uncertainty in climate change, the economic feasibility was found to be slightly reduced (7.5% increase in cost). Second, we re-evaluated the economic feasibility of the project using the ROA that considers two options, ‘delay’ and ‘abort’, under climate change uncertainty. Consequently, it was found that a significant amount of cost can be reduced through the selection of the ‘abort’ option. Though, the ROA indicated that the conduit project is not economically feasible, based on a sensitivity analysis conducted on the ROA, it was discovered that the project would be economically valuable if the current occurrence probability of extreme drought (approximately 1.43%) is increased by 1.2 times.

3.
HS22-D5-AM1-301-031 (HS22-A035)
 
Risk Assessment of Future Extreme Drought According to Climate Change Scenarios
Ji-Eun KIM1+, Si-Jung CHOI2, Jisoo YU1, Tae-Woong KIM1#
1 Hanyang University, South Korea, 2 Korea Institute of Civil Engineering and Building Technology, South Korea
#Corresponding author: twkim72@hanyang.ac.kr +Presenter

Recently, climate change has caused abnormal and extreme climatic phenomena in many parts of the world. Consequently, extreme drought is frequent in Korea, and the damage has reached a very serious level. Therefore, a probabilistic analysis method for future extreme drought risk is required to minimize damage from extreme drought. In recent years, even though various studies have been conducted for assessing the risk of extreme drought, we need to pay attention to the change of risk due to expected future extreme droughts compared with the present. This study quantitatively assessed future extreme drought risk considering severe drought characteristics according to 19 climate change scenarios. The drought characteristics, i.e., duration and severity, were estimated using the threshold level method, which are based on the concept that the drought is defined when precipitation is less than the climatological average. In order to solve the problem that the results are different when performing the univariate frequency analysis for the duration and severity, the bivariate frequency analysis was employed with the copula function so that the duration and the severity can be simultaneously analyzed. The drought risk was calculated based on the return period obtained by the bivariate frequency analysis and the risk of future extreme drought was analyzed and compared with the present maximum drought. As a result, the average durations of current and future are similar to 2.5 months. However, the average severity generally increased by more than 1.3 times from 130 ~ 150 mm to 190 ~ 210 mm. Therefore, it is expected that more severe drought will occur in the future compared to the present. In addition, drought that is larger than the current maximum drought event would occur in the future in about 20 events in most regions. Future extreme droughts above the current maximum drought have a risk of greater than 0.9, so the occurrence probability of extreme drought is high in the future.

Acknowledgement:

This work is supported by the Korea Agency for Infrastructure Technology Advancement (KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant 18AWMP-B083066-05) and the National Research Foundation (NRF- 2016R1D1A1A09918872).

4.
HS22-D5-AM1-301-032 (HS22-A055)
 
Fluvial and Pluvial Flood Risk Curve and its Future Changes in Urban Areas: A Case Study of the Shonai River Basin, Japan
Tomohiro TANAKA#+, Keiko KIYOHARA, Yasuto TACHIKAWA, Yutaka ICHIKAWA, Kazuaki YOROZU
Kyoto University, Japan
#Corresponding author: tanaka.tomohiro.65m@kyoto-u.jp +Presenter

Flood risk assessment plays more and more importance roles in a changing climate. As the most straightforward approach to flood risk assessment, a flood risk curve, which is the exceedance probability of flood damage, has been developed in a number of studies. Flood risk curves clearly show the effect of various types of countermeasures as the decrease of the exceedance probability of flood damage for both structural/non-structural countermeasures. 

Flood damage is caused by fluvial and pluvial flooding, especially in urban areas. The risk for these two types of flooding has been evaluated separately; probably because the major preparation is different, i.e. fluvial flooding is prevented by dam/levee construction while pluvial flooding is managed by improving sewage drainage system. Moreover, the dominant flood risk is assumed to be caused by fluvial flooding because it causes more severe damage than pluvial flooding once it occurs. On the other hand, it is also true that pluvial flooding tends to cause more frequent damage than fluvial flooding, and may not result in smaller economic loss than fluvial flooding because it can occur in a wider range. As the result, pluvial flooding does not necessarily have smaller risk than fluvial flooding.

For integrated flood risk assessment in urban areas, this study developed flood risk curves for both the fluvial and pluvial flooding in present and future climate conditions in the Shonai River Basin, Japan. Flood risk assessment requires the large number of heavy rainfall event data to realize the flood damage samples; therefore, we utilized 3,000 years rainfall data from d4PDF and converted them into flood damage by incorporating pluvial flood mechanism into a fluvial flood-inundation model. As the result, the difference of the impact of fluvial and pluvial flooding on urban flood risk and its future changes were evaluated.

5.
HS22-D5-AM1-301-033 (HS22-A059)
 
Estimation for Storm Surge Height and Inundation Area on Climate Change at Osaka Bay, Japan
Yoko SHIBUTANI1#+, Sota NAKAJO2, Sooyoul KIM3, Nobuhito MORI4, Hajime MASE4
1 Toyo Construction Co., Ltd., Japan, 2 Osaka City University, Japan, 3 Tottori University, Japan, 4 Kyoto University, Japan
#Corresponding author: yoko.shibutani@gmail.com +Presenter

Introduction

Recently, it is reported that tropical cyclones become more intense due to warmer sea surface conditions, resulting in a significant increase of storm surge risks. Therefore, it is important to estimate future changes of storm surge risks including intensity and return period under the future warmer climate condition. In this study, we carried out a sensitivity tests of storm surge simulations to climate change considering typhoon characteristics and sea level rises in the Osaka bay.

Methodology

A series of storm surge simulations was carried out using the coupled model of Surge, Wave and Tide (SuWAT) developed by Kim et al. (2008). The atmospheric forcing is given by a parametric typhoon model in this study. The SuWAT model employs the nesting scheme with the three leveled-domains from the resolutions of 7,290 m to 30 m. First, the track involved in worst surge levels in the Osaka Bay was estimated by shifting and rotating the track of 2nd Typhoon Muroto. Secondly, future changes of 2nd Typhoon Muroto were analyzed based on GCM outputs and sea level rise. Finally, the return period was estimated from a stochastic global tropical cyclone model (Nakajo et al., 2014).

Results

We estimated the possible maximum surge level by changing the central pressure depression and perturbed track of 2nd Typhoon Muroto for the Osaka Bay as a case study. Our results indicated that the possible maximum surge height is 1.0m higher than the 2nd Typhoon Muroto. The change of inundation area due to climate change will be increased about 160 % in comparison with the historical inundated area. In addition, we estimated that the return period of the possible maximum surge height was 21 years obtained using stochastic tropical cyclone model.

6.
HS22-D5-AM1-301-034 (HS22-A050)
 
Estimation of Public Preference for High-Tide Disaster Risk Reduction Under Uncertainty
Toshio FUJIMI#+
Kumamoto University, Japan
#Corresponding author: fujimi@kumamoto-u.ac.jp +Presenter

Public support is required to implement adaptation policies for natural disaster risks. Prediction of natural disaster risk often includes uncertainty due to lack of knowledge on underling mechanism and insufficient data. Thus, the predicted risk are represented as a bundle of probability distributions. This study reveals public preference for high-tide disaster risk reduction under uncertainty with contingent valuation method. For that purpose, we implemented web-survey for residents in Osaka bay area. Typhoon and inundation simulations provide 25 high-tide risk curves as uncertainty in each respondent living area. In the survey, respondents are presented with two types of high-tide risk curves: average risk curve and worst risk curves among 25 risk curves. Then, we ask their willingness to pay (WTP) for eliminating high-tide inundation risks with hypothetical full-cover insurance. Applying a contraction model, a decision model under uncertainty, to survey data, we estimate respondent’s attitude for uncertainty aversion. In addition, we evaluate how uncertainty aversion attitude affects WTP for high-tide risk reduction and examine how it relates with respondent socio-economic attributes.

7.
HS22-D5-AM1-301-035 (HS22-A018)
 
Development of Bias Correction Methods and of Extreme Values Assessment Technology
Toshikazu KITANO#+
Nagoya Institute of Technology, Japan
#Corresponding author: kitano@nitech.ac.jp +Presenter

Bias correction and assessing the extreme values plays an important role in climate change researches by numerical models. Especially when numerous ensemble members are available, its overwhelming amounts should be utilized for the numerical processing in statistical viewpoint. The dataset from d4PDF (database for Policy Decision-Making for Future Climate Change) consists of 50 members of 60 years duration for case of the past climate. Someone can use it by putting all daily precipitations of 50 members together into 3,000 years, and thus we would believe that the design rainfall of our target return periods, 50, 100 and 200 years are calculated and it would be a kind of interpolation without bothering the uncertainty. But the reality is not so easy. Rather, now we can handle the uncertainty with numerous ensemble members at the level of taming of the unpredictable extremes.

This research insists on the numerous ensemble members which gives the own result of estimation, and therefore we propose here a new framework of summarizing the results obtained by individual ensemble member. Modestly our target levels for return periods are treated as extrapolation for each ensemble member whose length is limited to be 60 years (obviously less than 100 or 200 years). Our framework employs the techniques that are found in ordinal MLE (Maximum Likelihood Estimation) and Bayesian approach by MCMC methods. One of the key ideas is to focus a certain relation between the error variance of return level and that of (log-transformed) return period, and its relation is so enough simple to be feasible for large amount of calculation. Another aspect of this framework shows that it can be examined whether the ordinal extreme value analysis of using the annual max. works in concordance with the properties of the 60 yrs' max. of each ensemble members, and if it is not concordant, the bias of the tail part of the probability distribution of heavy rainfall will be revealed in the statistical sense.

8.
HS22-D5-AM1-301-036 (HS22-A044)
 
A Comparison of Stochastic Extreme Downscaling Models for an Assessment of Changes in Rainfall Intensity-Duration-Frequency Curves over South Korea
Hyun-Han KWON#+, Yong-Tak KIM, Huy NGUYEN DINH, Hong-Geun CHOI
Chonbuk National University, South Korea
#Corresponding author: hkwon@jbnu.ac.kr +Presenter

We have always faced climate risks associated with climate change and climate variability, and increasing attention has been paid to the influence of climate change on the extreme rainfall. Extreme rainfall has become more frequent over the Korean peninsula in recent years, causing serious damages. Climate change projections for precipitation are in general provided at daily time step. However, sub-daily precipitation data is necessarily required for hydrologic design and management, especially for the estimation of design rainfalls. Thus, we developed three different stochastic approaches to downscale the daily maximum rainfalls to the sub-hourly maximum series in a changing climate . First, we focus on exploring a Copula function in a Bayesian inference framework to estimate rainfall IDF curves. Second, we investigate a Local-Regional Scaling-Invariant approach in a Bayesian inference framework to estimate regional IDF curves. Third, a four parameter Best distribution based nonstationary frequency analysis model is finally introduced. This study explores how likely it is that design rainfall for some stations in South Korea would shift under climate change based on Representative Concentration Pathways (RCPs) scenarios under the CORDEX experiments.

Acknowledgement

This work is supported by the Korea Agency for Infrastructure Technology Advancement(KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant 18AWMP-B083066-05).

HS22 - Climate Change Risk Assessment and Adaptation on Water-related Disaster and Water Resources in Asia and the Pacific
Friday, June 08, 2018 | 301 | 11:00-12:30
1.
HS22-D5-AM2-301-037 (HS22-A048)
 
Assessing Quantile Mapping Method for GCM Outputs Bias Correction
Ke-Sheng CHENG#, Bo-Yu CHEN+
National Taiwan University, Taiwan
#Corresponding author: rslab@ntu.edu.tw +Presenter

Bias correction is an essential work for climate change impact studies. Many methods for bias correction exist and quantile mapping appears to be most widely adopted. The strength of quantile mapping bias correction (QMBC) method lies in its capability of preserving statistical moments of different orders based on observed data of the baseline period. However, for the projection period, empirical cumulative distribution function (ECDF) of the observed variable of interest does not exist and it is customary to use ECDF of the baseline period instead. Whether such practice can still preserve the unknown statistical moments of the projection period requires further investigation. In addition, it is often assumed that biases relative to historical observations will be constant in the projection period when conducting bias correction. In this study, we assessed the performance of quantile mapping bias correction method through stochastic simulation. It was found that while the QMBC method could preserve the mean of the variable of interest in the projection period, it generally underestimated the extremes. Detailed assessments, their implications, and possible improvements will be presented at the conference.

2.
HS22-D5-AM2-301-038 (HS22-A053)
 
Dual Window Bias Correction for Hourly Precipitation Projected by Super Ensemble Experiments
Satoshi WATANABE#+
The University of Tokyo, Japan
#Corresponding author: stswata@sogo.t.u-tokyo.ac.jp +Presenter

In this study, we present a bias correction method for hourly precipitation projected by super ensemble experiments. Future hourly precipitation in Japan for over 6000 years was acquired from the database for policy decision-making for future climate change (d4PDF), and corrected using automated meteorological data acquisition system (AMeDAS) as the reference data. The advantages of this method are; 1) the merits of typical bias correction method, bias-preserving and trend preserving, are combined; 2) comparison between model and reference considering widows centered around time and rank improves the robustness of bias correction. Bias correction methods are classified into two types, and they both have their advantages and disadvantages. An improvement of previous method to use both advantage was required. The conventional comparison method, quantile based mapping, was found to lack in robustness because it compares quantile value individually, and thus required improvement.

The method was validated by applying the method to historical datasets of two river basins, in Japan, the Omono River basin and the Hiji River basin. The result indicates that the corrected extreme precipitation projected by super ensemble experiments can reproduce well the design precipitation that is parametrically estimated from historical reference datasets using an extreme value statistical method. In particular, corrected data for once in 150 years precipitation in Omono River basin and once in a 100 years in Hiji River that is 258mm/2day and 340 mm/2day is 252 and 349, respectively. This indicates that bias corrected extreme precipitation obtained from super ensemble simulations using order statistics can represent extreme precipitation estimated from reference data using an extreme value statistical method.

3.
HS22-D5-AM2-301-039 (HS22-A002)
 
Impact Assessment of Climate Change on Coastal Hazards in East Asia
Nobuhito MORI#+
Kyoto University, Japan
#Corresponding author: mori@oceanwave.jp +Presenter

Understanding future changes of ocean waves and storm surges is important for assessing and mitigating the impact of climate on coastal, marine and ocean environments and on engineering problems. The latest research results of climate change impacts on coastal hazards will be summarized in this presentation. First, future changes of wave climate and storm surges based on GCM ensemble experiments are summarized. Second, the applications of coastal hazard projections to coastal structures and beach profiles are summarized as a series of climate impact assessment projects. There are clear increases in extreme values of wave heights and storm surges in the tropical cyclone dominant regions around the middle latitudes of the Western North Pacific including Japan. 

4.
HS22-D5-AM2-301-040 (HS22-A019)
 
Potential Wave-Attacked Risk Maps for Extreme Typhoon Events Along the Coast of Taiwan
Hung-Ju SHIH1#, Ting-Yu LIANG1, Chih-Hsin CHANG1, Wei-Bo CHEN1, Lien-Kwei CHIEN2+
1 National Science and Technology Center for Disaster Reduction, Taiwan, 2 National Taiwan Ocean University, Taiwan
#Corresponding author: shihruby713@ncdr.nat.gov.tw +Presenter

Every year, the coastal areas of Taiwan were attacked by typhoon-induced giant waves. The roaring waves caused by typhoon Meranti (2016) even destroyed the lighthouse of a fishing port on the southwestern coast of Taiwan. Therefore, there has been increased interest in generating potential wave-attacked risk maps for extreme typhoon events along the coast of Taiwan. In this study, the highest intensity typhoon (HIT) for each category (the Central Weather Bureau in Taiwan classified all typhoons’ tracks into 9 categories) from 1977 to 2016 were selected. A state-of-the-art high-resolution unstructured-grid fully coupled tide-surge-wave model and 10-meter winds above sea level from dynamically downscaled WRF (Weather Research and Forecasting) model were utilized to reproduce the distribution of maximum significant wave height (DMSWH) for 9 typhoon events. Each DMSWH was classified as one of 5 levels and was employed to create the maximum possible potential wave-attacked risk maps for extreme typhoon events. Our results demonstrate that the northern and the entire eastern coasts of Taiwan tended to be attacked by violent waves (7 m ≤ VW < 11.5 m) for a coastline length of 236.4 km and roaring waves (RW ≥ 11.5 m) for a coastline length of 298.1 km when the HIT conditions were present. The potential wave-attacked risk maps for extreme typhoon events created in this study are useful for designing and deploying coastal infrastructure such as seawalls and tetrapods. Additionally, they could be used to prepare a detailed evacuation plan for associated wave attack emergencies due to the approach of typhoons.

5.
HS22-D5-AM2-301-041 (HS22-A041)
 
Downscaling of Coastal Current in the Eastern Japan with Included Freshwater Impact
Josko TROSELJ1#+, Yuki IMAI1, Junichi NINOMIYA2, Nobuhito MORI1
1 Kyoto University, Japan, 2 Kanazawa University, Japan
#Corresponding author: josko.troselj@gmail.com +Presenter

Introduction and Objective Global mean sea level, surface temperatures and extreme precipitation amounts are expected to increase in 21st century under the climate change impact. It will especially affect coastal zones which are sensitive to all of these factors because extreme freshwater inflow, inundation levels, erosion as well as fishery industry nearshore are expected to be largely impacted.
 Objective of the study is developing methods for downscaling of coastal current system for Ibaraki prefecture in Japan from 10 km scale parent dataset to related projections of 200 m scale with included freshwater impact from three major rivers. Its purpose is to formulate techniques for providing fine scale ocean circulation reanalysis and future projection which can be used for adaptation of countermeasures for climate change impact assessment. With using downscaling approach, our main focus is natural variability of physical processes of the ocean rather than their long term trend.

Methods We used lateral boundary conditions of velocity, temperature and salinity from FORA-WNP30 parent dataset (10 km scale) downscaled by COAWST model using 3 domain nesting with 2 km, 600 m and 200 m scales respectively, for targeted reanalysis period of 2000. 
 Surface boundary conditions of wind speed, sea level pressure, air temperature, precipitation, shortwave radiation flux and cloud fraction are used from JRA-55 dataset, while forcing conditions are used as observed temperature and discharge with constant salinity of 0.5 PSU from Tone, Naka and Kuji rivers.

Discussions

We compared our modeled results with observed surface temperature data from Hasaki point (from PARI) and coastal zone near Ibaraki (from JMA). We found noticeable freshwater impact at Hasaki point and improved results of surface temperature from the parent dataset by downscaling both at Hasaki point and at the coastal zone near Ibaraki.

Keywords: downscaling, coastal current, freshwater impact, natural variability, climate change, COAWST.

6.
HS22-D5-AM2-301-042 (HS22-A037)
 
High-Resolution Wave Climate Projection for Northwestern Atlantic and Coastal Eastern USA
Adrean WEBB#+, Tomoya SHIMURA, Nobuhito MORI
Kyoto University, Japan
#Corresponding author: adrean.webb@gmail.com +Presenter

A high-resolution wave climate projection for the northwestern Atlantic Ocean has been conducted by DPRI (Kyoto University) to help assess possible regional impacts due to global climate change. The spectral wave model NOAA WAVEWATCH III is utilized with three coupled (two-way) grids to resolve the northwestern Atlantic and coastal eastern USA at approximately 21 km and 7 km (respectively), and covers the periods 1979-2003 (historic) and 2075-2099 (future). Hourly wind field forcings are provided by a high-resolution AGCM (MRI-AGCM 3.2S; 21 km) and allow for better modeling of large storm events (important for extreme event statistics). Climatological (25-year) significant wave height differences between future and historical periods indicate a seesaw effect will occur, with a northward decrease approaching 1 m in areas north of approximately 23 deg latitude in the Atlantic Ocean, and a southward increase approaching 0.5 m in the Caribbean Sea. Here, model validation, wave climate projection, and comparison with other relevant studies will be discussed.

Poster Presentations

  HS22-D2-PM1-P-043 (HS22-A001)
 
Future Dispersal and Connectivity of World’s Northernmost Coral Reefs
Shintaro TAKAO1#+, Hiroshi KURODA2, Hiroya YAMANO3, Masahiko FUJII4, Yasuhiro YAMANAKA4
1 National Institute of Polar Research, Japan, 2 Fisherie Research Agency, Japan, 3 National institute for Environmental Science, Japan, 4 Hokkaido University, Japan
#Corresponding author: takao.shintaro@nipr.ac.jp +Presenter

Biogeographical shifts in marine organisms driven by climate change have recognized in the world ocean. Changes in ocean temperatures and currents are expected to affect the latitudinal limits and distributions of coral species with changes in several ecological aspects, including larval dispersal and connectivity. However, understanding of these changes in northernmost corals, as the touchstone of future coral distributions, is extremely limited. Using an individual-based biophysical dispersal model driven by an eddy-resolving ocean model, we simulated dispersal and connectivity of modeled coral larvae under historical and two global warming scenarios (RCP2.6 and 8.5), to study these potential changes in the highest latitude coral reef areas. Results under the historical simulation suggest that Ryukyu Islands, where extending the southern Japanese main island (Kyushu) to northeastern Taiwan, would be important sources of coral larvae for islands south of Kyushu. Furthermore, coastal areas of Taiwan and China may be remote sources for the western coastal areas of Kyushu and Tsushima Island, where world’s northernmost coral reefs exist. Under the RCP8.5 scenario, on the other hand, an enhanced countercurrent in the south of Ryukyu Islands would limit the poleward expansion of corals from Ryukyu Islands to islands south of Kyushu, although ocean warming will expect to increase local retention of larvae. Our results of future coral dispersal and connectivity will help develop strategies for coral reef management and conservation among nations and marine protected area designs in the world’s northernmost corals.

  HS22-D2-PM1-P-044 (HS22-A005)
 
Application of Flood Vulnerability Index for Analyzing Safety Change of Levee according to Climate Change
Hoo Sang LEE1#+, Jae Joon LEE1, Jun-Haeng HEO2, Sung Ho LEE1
1 Kumoh National Institute of Technology, South Korea, 2 Yonsei University, South Korea
#Corresponding author: toyou3045@naver.com +Presenter

In this study, a new technique for evaluating the flood vulnerability of river banks is proposed. For this purpose, flood quantities of the basin were estimated based on the future climate change scenarios and the infiltration stability was evaluated by analyzing the infiltration behavior using SEEP/W which is a 2D groundwater infiltration model of the levee. The size of the river levee was investigated. The size of river levee was investigated by selecting the target area. The safety factor of the levee was analyzed considering the current flood level of the levee and the flood level considering the climate change. The factor needed to analyze the levee vulnerability was derived. We analyzed the vulnerability of the levee considering the change of the levee level according to the climate change scenarios. Levee Flood Vulnerability Index(LFVI) were used to evaluate the vulnerability of the levee.

"This work is supported by the Korea Agency for Infrastructure Technology Advancement(KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant 18AWMP-B083066-05)."

  HS22-D2-PM1-P-045 (HS22-A015)
 
Drought Early Warning and Preparedness: A Case Study in Southern Taiwan
Hung-Wei TSENG1+, Chen-Min KUO1, Tao-Chang YANG1, Pao-Shan YU2#
1 National Cheng Kung University, Taiwan, 2 -, Taiwan
#Corresponding author: yups@mail.ncku.edu.tw +Presenter

This study aims to establish a drought early warning and preparedness framework for a multi-objective reservoir, Tsengwen Reservoir, and its downstream area in southern Taiwan. Since water demands (i.e., agricultural, domestic and industrial uses) in the downstream area of reservoir highly relies on the water supply of reservoir, the area is extremely vulnerable to drought. Increasing the resilience of the water supply system and mitigating the potential impact of droughts in the area are very crucial. To this end, a drought early warning and preparedness framework based on empirical standardized drought indexes (ESDI) is proposed in the study. Many drought indexes such as standardized precipitation index, standardized reservoir storage index and standardized streamflow index have been used for defining and monitoring droughts. Following the concepts of the previous literature, the study involved different hydrological variables (e.g., rainfall, reservoir inflow and storage) as the basis of ESDI to characterize and monitor drought condition for the reservoir. The study analyzed the relationship between the time series of ESDI and historical drought events (water shortage) to decide the critical threshold value (CTV) of ESDI. The CTV of ESDI is considered as a trigger for drought early warning and preparedness. Once the value of ESDI reaches the CTV, the actions of providing back-up water supply and water rationing can be immediately adopted for easing the impact of subsequent drought events. A comparison of the drought risk with and without early warning and preparedness will be made to investigate the performance of proposed method in risk mitigation.

  HS22-D2-PM1-P-046 (HS22-A016)
 
Evaluation of Extreme Rainfall Under Climate Change Scenario in Korea Peninsula
Minsung KWON1+, Jae-Hyun AHN2#
1 Urban Risk Management Research Center, South Korea, 2 Seokyeong University, South Korea
#Corresponding author: wrr@skuniv.ac.kr +Presenter

This study evaluated the risk of extreme rainfall under climate change in Korea Peninsula. Rainfall stations for the study are 88 stations that including 61 stations in South Korea and 27 in North Korea, and Representative Concentration Pathways (RCPs) scenarios produced by Korea Meteorological Administration was used. The scenario is consist of 4 types (RCP 2.6, RCP 4.5, RCP 6.0, RCP 8.0) according to radiative forcing. The risk of future extreme rainfall was calculated by its return period which was estimated by probability distribution of observed rainfall (1981-2010, SP0). The reason for applying this method is that it is important to express the risk of future extreme rainfall on the basis of a current characteristic of rainfall. RCP scenario was divided into three periods (2011-2040 (SP1), 2041-2070 (SP2), and 2071-2100 (SP3)) to compare changes over time. Ratios of rainfall stations with a return period over 300-year in SP1, SP2, and SP3 are 14%, 28%, and 35%. In the return period over 300-year, ratios of it over 1,000-year were very high as SP1 64%, SP2 63%, and SP3 69%.

Acknowledgement:

This work is supported by the Korea Agency for Infrastructure Technology Advancement (KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant 18AWMP-B083066-05).

  HS22-D2-PM1-P-047 (HS22-A039)
 
Projection of Future Change in Storm Surges by Artificial Neural Network and d4PDF
Yuji ARAKI1+, Tomohiro YASUDA1#, Nobuhito MORI2, Sota NAKAJO3
1 Kansai University, Japan, 2 Kyoto University, Japan, 3 Osaka City University, Japan
#Corresponding author: tomo@oceanwave.jp +Presenter

Projection of future change in storm surges needs a lot of data though number of observations of actual storm surges are limited. To solve this problem our group has developed stochastic tropical cyclone model (STM) to generate synthetic storms. Although STM generates tens of thousands tropical cyclones (TCs), simulation cost of storm surges by a numerical model is still high. To reduce the simulation cost this study aims to develop Artificial Neural Network (ANN) to project future change in storm surges statistically. 

Storm surge projection ANN uses TCs information as input data and outputs storm surge height. Input data for ANN are distance from a target point, central pressure, approach angle, and moving speed of TCs. Target area is Ise Bay, Japan and the number of TCs which passes there are about 700 in 1000 years of STM output. Nonlinear shallow water model uses the 700 storm tracks and simulates storm surges. To improve accuracy of ANN, we examine best combination of ratio of training data, number of neurons in middle layer, and lead time. We also compare accuracy in four input conditions: stationary input (SI), timeline input (TLI), conditional stationary input (CSI), conditional timeline input (CTLI).

Best input combinations were found for each conditions, e.g. 70% of training ratio, 5 neurons, and 1-hour lead time for the case of SI. CTLI showed best performance in verifications of comparison with observed storm surges over 1.2 m. The RMSE was less than 0.3m of CTLI whereas other conditions’ errors were 0.3 to 0.6 m. We used the d4PDF (database for Policy Decision making for Future climate change) which projects future climate in 4-degree warmer world by 6000-year ensemble simulations for present and future climate. To project future change in storm surge we applied developed ANN to d4PDF with best input condition.

  HS22-D2-PM1-P-048 (HS22-A043)
 
Intercomparison of Bias Correction Methods for Runoff Generation Outputs from Land Surface Models at the Chao Phraya River Basin
Teerawat RAM-INDRA, Patinya HANITTINAN#+, Yasuto TACHIKAWA, Yutaka ICHIKAWA, Kazuaki YOROZU
Kyoto University, Japan
#Corresponding author: hanittinan.patinya.5n@kyoto-u.ac.jp +Presenter

We discuss how much the bias in total runoff generation (ROF) output calculated from a widely-used Simple Biosphere Model (SiB) can be alleviated by conducting bias correction techniques in the present-day climate (1951-2007) at the Chao Phraya River Basin. The gridded observation data, APHRODITE was used to drive the model simulation. The simulated data were then compared with the “reference” runoff output from the Simple Biosphere Model with Urban Canopy (SiBUC) to evaluate the model bias during the calibration (1951-2000) and the verification periods (2001-2007). In this study, we adopt two bias correction approaches, e.g., quantile mapping and multiplicative shift to address the systemic errors in mean, standard deviation and frequency of the generated runoff from SiB.

  HS22-D2-PM1-P-049 (HS22-A046)
 
Statistical Downscaling of AGCM60km Precipitation Based on Spatial Correlation of AGCM20km Output
Sunmin KIM1#+, Yasuto TACHIKAWA2, Eiichi NAKAKITA2
1 -, Japan, 2 Kyoto University, Japan
#Corresponding author: kim.sunmin.6x@kyoto-u.ac.jp +Presenter

The Japan Meteorological Agency (JMA) and the Meteorological Research Institute (MRI) of Japan have developed a couple of atmospheric general circulation models (AGCM) with 20 km and 60 km horizontal resolutions. If the number of the AGCM20km ensemble outputs is as many as the number of the AGCM60km outputs, such as more than 100 ensemble outputs from the database for Policy Decision making for Future climate change (d4PDF) experiments (Mizuta et al., 2016), the applicability of the model outputs in variant impact assessment research will be drastically increased. 

In this study, a statistical downscaling method based on regressing precipitation data is only introduced and applied to 60-km resolution Atmospheric General Circulation Model (AGCM60km) output for daily precipitation. The method utilizes a regression domain with a 3×3 60-km grid, and the downscaling target is 3×3 20-km grids in the center of the regression domain (Kim et al., 2017). By shifting the regression domain one grid by one grid in 60-km resolution, the same form of regression model, but different regression coefficients for each 20-km grid, can be applied to all the downscaling target areas. Based on application tests for the Asian Monsoon region, the statistical downscaling algorithm shows extremely effective results with a certain pattern of regression error. The monthly based downscaled results from AGCM60km output shows a rather good match to the monthly mean precipitation amount of AGCM20km. The downscaled results also show a plausible mimic to the AGCM20km output in the frequency of daily precipitation amounts; however, the results showed noticeable limitations in simulating low rainfall amounts (e.g., less than 5 mm d-1), especially on land.

  HS22-D2-PM1-P-050 (HS22-A058)
 
Climate Change Impacts on Low Flow Regime in the Context of the Variation of Actual Evapotranspiration
Hoyoung SUN+, Boosik KANG#
Dankook University, South Korea
#Corresponding author: bskang@dankook.ac.kr +Presenter

The climate change impacts on hydrological flow regimes from mean values to high and low extremes are critical issues for adaptive water management. This study predicted the future change in low-flow under the climate change scenarios, which is the basis for the assessment of water supply reliability. The hydrometeorological observation data during reference period (1974~2000) and the five available climate change model (ACCESS1.3, CanESM2, CNRM-CM5, GFDL-ESM2G, HadGEM2-AO) of CMIP5 under the IPCC RCP 4.5/8.5 scenario during future period (2011~2100) of Soyang Dam basins in Han River, the Republic of Korea were used. The uncertainty in ensemble estimation is reduced through the Bayesian Model Averaging technique after correcting systematic biases of the climate change data using nonstationary quantile mapping. In the future period, precipitation increase at a rate of 1.89mm/year under RCP 4.5 and 3.49mm/year under RCP 8.5. Actual evapotranspiration increase at a rate of 1.17mm/year under RCP 4.5 and 1.89mm/year under RCP 8.5. The use of CMD(Catchment Moisture Deficit) non-linear module of IHACRES model was used to improve the estimation process of effective rainfall by directly use of the actual evapotranspiration. The R-squared between the observed and simulated flow in the reference period is 0.79. The projected stream flow increase at a rate of 1.94MCM/year under RCP 4.5 and 4.32MCM/year under RCP 8.5. The projected drought flow by flow-duration decrease at a rate of 0.02MCM/year under RCP 4.5 and 0.03MCM/year under RCP 8.5. Increased precipitation did not lead to increase in drought flow because of the increased actual evapotranspiration, despite increased streamflow under RCP 4.5/8.5. The reduced drought flow in the future can threaten the sustainability of instream flow or irrigation water supply.