第119回総合防災セミナー(9/29)
- セミナー
| 開催期間 | 2026年9月29日（火） 15:30 ～ 17:00 |
|---|---|
| 場所 | 宇治キャンパス本館E517D および オンライン |
| 対象 | 研究者、学生、一般 |
9月29日（火）に第119回総合防災セミナーを開催します。
皆様のご参加をお待ちしています。本セミナーは、ハイブリッドかつ英語で行います。
[Date & Time]
9月29日（火）15:30-17:00
9月29日（火）15:30-17:00
[Venue]
宇治キャンパス本館E517D および オンライン
※オンラインでご参加の方は9月28日（月）
登録フォーム：https://forms.gle/
[Speaker]
Dr. Md. Touhidul Islam
Dr. Md. Touhidul Islam
Bangladesh Agricultural University
[Title]
Hydrological Modeling and Deep Learning Approaches for Flood and Drought Prediction in Bangladesh River Basins: From Irrigation Water Management to Ongoing Research at DPRI
[Abstract]
Water-related disasters, particularly floods and droughts, pose severe challenges to agriculture and water resources management in Bangladesh. This presentation gives a brief overview of my research as a whole, from my past work to my present research here at DPRI, addressing these challenges through hydrological modeling, remote sensing, and machine learning. In earlier work, I examined the impacts of climate change on irrigation water requirements and agricultural water management under CMIP5 and CMIP6 scenarios, with a focus on drought and flood risks facing major crops. Since joining DPRI as a JSPS postdoctoral fellow under the supervision of Sayama sensei, I have evaluated the long-term performance of multiple satellite-based rainfall products across the whole of Bangladesh, assessing their reliability for hydrological applications in a data-scarce region. Building on this, I am now applying the RRI model to continuously simulate joint flood and drought conditions, targeting both high-flow and low-flow prediction for major Bangladeshi river basins. In parallel, I am developing advanced deep learning models for continuous discharge prediction in these basins. This talk will present key findings from these studies, share the current progress of my research at DPRI, and outline future directions for improving water-related disaster prediction in Bangladesh/Japan.
Water-related disasters, particularly floods and droughts, pose severe challenges to agriculture and water resources management in Bangladesh. This presentation gives a brief overview of my research as a whole, from my past work to my present research here at DPRI, addressing these challenges through hydrological modeling, remote sensing, and machine learning. In earlier work, I examined the impacts of climate change on irrigation water requirements and agricultural water management under CMIP5 and CMIP6 scenarios, with a focus on drought and flood risks facing major crops. Since joining DPRI as a JSPS postdoctoral fellow under the supervision of Sayama sensei, I have evaluated the long-term performance of multiple satellite-based rainfall products across the whole of Bangladesh, assessing their reliability for hydrological applications in a data-scarce region. Building on this, I am now applying the RRI model to continuously simulate joint flood and drought conditions, targeting both high-flow and low-flow prediction for major Bangladeshi river basins. In parallel, I am developing advanced deep learning models for continuous discharge prediction in these basins. This talk will present key findings from these studies, share the current progress of my research at DPRI, and outline future directions for improving water-related disaster prediction in Bangladesh/Japan.
[Biography]
ISLAM Md Touhidul is an Associate Professor in the Department of Irrigation and Water Management, Bangladesh Agricultural University (BAU), Bangladesh, and currently a JSPS Standard Postdoctoral Fellow at the Disaster Prevention Research Institute (DPRI), Kyoto University, under the supervision of Prof. Takahiro Sayama. He completed his PhD at Okayama University in 2022, receiving the Dean’s Award for Outstanding PhD Graduate. His research focuses on hydrological modeling, agricultural water management, climate change impact assessment, GIS-based spatial analysis, and machine learning and deep learning applications for hydrometeorological prediction. He values collaboration with researchers across different relevant fields to address problems from multiple perspectives. He has published 73 articles to date, including 15 in Q1 and 23 in Q2 journals. He has received several recognitions, including the JSPS Postdoctoral Fellowship, the Top Ten Researcher Award at BAU (Scopus ranking, 2025), and the Best Publication Award. He is currently living in Uji, Kyoto, with his wife and young son (5 years old) during his fellowship at DPRI.