Please let me introduce myself, my name is Albert,PhD, from Semarang University
(USM), Semarang, Indonesia. I am interested with your topic.
My questions are:
1. How important this research ?
2. Why are using the LSTM networks and GTS-based visualisation ?
Replies:
I am Dr. Nadir La Djamudi from Universitas Muhammadiyah Buton, Baubau City.
Although I am a linguist, I am very interested in your research topic, ^Multivariate
Time-Series Flood Prediction Using LSTM Networks and GIS-Based Visualization of
Meteorological Data from BMKG,^ because I live in Baubau City.
My question is: What are the contributions of your research findings for Baubau City,
both at present and in the future?
The findings of this research provide immediate contributions for Baubau City by delivering a data-driven flood forecasting and spatial visualization framework that supports early warning and localized risk identification. In the future, the proposed integration of deep learning and GIS can serve as a scalable decision-support tool for urban flood management, enabling improved planning, infrastructure adaptation, and disaster preparedness as hydrometeorological risks increase.
I am Dr. Nadir La Djamudi from Universitas Muhammadiyah Buton, Baubau City.
Although I am a linguist, I am very interested in your research topic, ^Multivariate
Time-Series Flood Prediction Using LSTM Networks and GIS-Based Visualization of
Meteorological Data from BMKG,^ because I live in Baubau City.
My question is: What are the contributions of your research findings for Baubau City,
both at present and in the future?
The findings of this research provide immediate contributions for Baubau City by delivering a data-driven flood forecasting and spatial visualization framework that supports early warning and localized risk identification. In the future, the proposed integration of deep learning and GIS can serve as a scalable decision-support tool for urban flood management, enabling improved planning, infrastructure adaptation, and disaster preparedness as hydrometeorological risks increase.
I am Dr. Nadir La Djamudi from Universitas Muhammadiyah Buton, Baubau City.
Although I am a linguist, I am very interested in your research topic, ^Multivariate
Time-Series Flood Prediction Using LSTM Networks and GIS-Based Visualization of
Meteorological Data from BMKG,^ because I live in Baubau City.
My question is: What are the contributions of your research findings for Baubau City,
both at present and in the future?
The findings of this research provide immediate contributions for Baubau City by delivering a data-driven flood forecasting and spatial visualization framework that supports early warning and localized risk identification. In the future, the proposed integration of deep learning and GIS can serve as a scalable decision-support tool for urban flood management, enabling improved planning, infrastructure adaptation, and disaster preparedness as hydrometeorological risks increase.
Why was short-term rainfall chosen as the primary indicator of flood potential, and not river level, discharge, or other hydrological combinations?
Replies:
Short-term rainfall was chosen because it is a major trigger for flooding in coastal urban areas such as Baubau City, where intense rainfall rapidly generates surface runoff and urban inundation. Rainfall provides an earlier and more direct signal of potential flooding compared to river levels or discharge, which typically respond with a temporal delay after a rainfall event. Furthermore, BMKG rainfall data is consistently available with high temporal continuity, making it suitable for short-term forecasting and early warning purposes. River levels and discharge are recognized as important complementary indicators and will be considered in future multi-source hydrological modeling.
How does the integration of multivariate time-series data in LSTM networks improve
flood prediction accuracy, and in what ways does GIS-based visualization of BMKG
meteorological data enhance decision-making for early flood warning and disaster
mitigation?
Replies:
GIS-based visualization enhances early flood warning and disaster mitigation by transforming BMKG meteorological data and model predictions into spatially explicit and easily interpretable maps. By integrating forecasting results within a geographic context, decision-makers can rapidly identify high-risk locations, assess spatial variability, and prioritize response actions. This spatial perspective improves situational awareness, supports more targeted mitigation strategies, and strengthens communication between technical analysts and local disaster management authorities.
