Melola Discovery

← Melola Discovery

Akses terbuka Jurnal SINTA 1

A Dual-Model Spatiotemporal Flood Prediction System Using Sentinel-1 SAR Imagery and Meteorological Data: A Case Study in Palembang, Indonesia

Journal of Robotics and Control (JRC) · Vol. 6 No. 6 (2026) · hlm. 3198–3213

DOI
10.18196/jrc.v6i6.27546
ISSN
2715-5072, 2715-5056
Terbit
2026-02-14
Penerbit
Universitas Muhammadiyah Yogyakarta
Bahasa
eng
Lisensi
creativecommons.org/licenses/by-sa/4.0

Abstrak

Floods remain one of the most destructive and frequent natural hazards, especially in urban river basins like Palembang, Indonesia. Improving early warning systems through the integration of radar and meteorological data has become increasingly feasible with advances in remote sensing and machine learning. The research contribution is the development of a dual-model spatiotemporal prediction framework that combines Sentinel-1 SAR imagery and meteorological data using eXtreme Gradient Boosting (XGBoost) for spatial classification and Long Short-Term Memory (LSTM) for temporal forecasting. Radar backscatter values were extracted from Sentinel-1 Single Look Complex (SLC) data and processed into one-kilometer resolution grids using SNAP and QGIS. These were merged with weather variables precipitation, humidity, wind speed, and solar radiation sourced from local meteorological stations. The XGBoost model achieved a precision of 98.3%, accuracy of 99.94%, recall of 97.5%, and F1-score of 97.9%, with SHAP analysis identifying rainfall and wind speed as dominant flood predictors. Spatial predictions aligned closely with historically flood-prone areas along the Musi River. In contrast, the LSTM model, despite forecasting floods up to 12 days in advance with average accuracy of 91.6%, suffered from class imbalance, resulting in a recall of only 22.9% and precision of 36.3%, limiting its applicability for real-time early warning. These findings demonstrate that while spatial classification using radar and weather data is highly effective, temporal forecasting remains challenged by data imbalance and uneven class distribution. Future research should explore cost-sensitive learning, uncertainty quantification, and real-time validation to enhance the system’s operational reliability.

Kata kunci

Jaringan kolaborasi para penulis

40 penulis · 94 tautan kepenulisan bersama · 23 lintas organisasi · menampilkan 40 yang paling banyak terhubung

Arahkan kursor atau ketuk sebuah nama untuk menyorot rekan penulisnya. Klik dua kali untuk membuka halaman penulis.

  • Politeknik Negeri Sriwijaya
  • Padjadjaran University
  • Bandung Institute of Technology
  • Poltekkes Kemenkes Surabaya
  • Sriwijaya University
  • Universiti Brunei Darussalam
  • State University of Semarang
  • Organisasi lain
  • Afiliasi tidak diketahui
  • Organisasi berbeda
  • Organisasi sama
  • Afiliasi tidak diketahui
  • Garis lebih tebal = lebih banyak artikel bersama · lingkaran lebih besar = lebih banyak artikel
Tampilkan sebagai tabel
Rekan penulisOrganisasiArtikel bersama
Dewi Liesnoor SetyowatiState University of Semarang3
Triwiyanto TriwiyantoPoltekkes Kemenkes Surabaya3
Ahmad TaqwaPoliteknik Negeri Sriwijaya2
Siti NurmainiSriwijaya University2
Irsyadi YaniSriwijaya University2
Ade SilviaBandung Institute of Technology2
Sriyanto SriyantoAfiliasi tidak diketahui2
Ananto AjiAfiliasi tidak diketahui2
Heri TjahjonoAfiliasi tidak diketahui2
Erni SuhariniAfiliasi tidak diketahui2
Wahyu SetyaningsihPoltekkes Kemenkes Malang2
Sari LuthfiyahPoltekkes Kemenkes Surabaya2
Abdussalam Ali AhmedBani-Waleed University2
Toshihiko NoguchiShizuoka University2
Wahyu LestariUniversity of Pembangunan Nasional Veteran Jawa Timur2
Grzegorz KrólczykOpole University of Technology2
M. ArdiansyahPoliteknik Negeri Sriwijaya1
Nur HopipahPoliteknik Negeri Sriwijaya1
Mohammad FadhliPoliteknik Negeri Sriwijaya1
Hanny HafiarPadjadjaran University1
Syauqy LukmanAustralian National University1
Putri LimiliaPadjadjaran University1
Kholidil AminPadjadjaran University1
Ahmad FaroziAfiliasi tidak diketahui1
Apik Budi SantosoAfiliasi tidak diketahui1
agus wahit rohmanAfiliasi tidak diketahui1
Tukidi TukidiAfiliasi tidak diketahui1
Agus PurnomoAfiliasi tidak diketahui1
Bregas Widya PratamaAfiliasi tidak diketahui1
Dini Yuniar SafitriAfiliasi tidak diketahui1
Tjaturahono Budi SanjotoAfiliasi tidak diketahui1
Edi PranotoAfiliasi tidak diketahui1
Doni PrayogoAfiliasi tidak diketahui1
Nurul SafaridaAfiliasi tidak diketahui1

Afiliasi berasal dari OpenAlex; metadata OAI-PMH jurnal tidak memuatnya. Penulis tanpa ID OpenAlex dicocokkan berdasarkan nama, sehingga dua orang bisa tergabung atau satu orang bisa terpisah.

Sumber metadata: OAI-PMH (metadata jurnal) https://journal.umy.ac.id/index.php/jrc/oai, pertama di-harvest 9 Nov 2026, terakhir terlihat 16 Nov 2026.
Metadata asli (oai_dc)
<oai_dc:dc xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/  http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
	<dc:title xml:lang="en-US">A Dual-Model Spatiotemporal Flood Prediction System Using Sentinel-1 SAR Imagery and Meteorological Data: A Case Study in Palembang, Indonesia</dc:title>
	<dc:creator>Handayani, Ade Silvia</dc:creator>
	<dc:creator>Aritonang, Clara Silvia Rotua</dc:creator>
	<dc:creator>Suroso, Suroso</dc:creator>
	<dc:creator>Husni, Nyayu Latifah</dc:creator>
	<dc:creator>Caesarendra, Wahyu</dc:creator>
	<dc:creator>Novianti, Leni</dc:creator>
	<dc:subject xml:lang="en-US">Flood Prediction</dc:subject>
	<dc:subject xml:lang="en-US">Spatiotemporal Analysis</dc:subject>
	<dc:subject xml:lang="en-US">Sentinel-1</dc:subject>
	<dc:subject xml:lang="en-US">Weather Data</dc:subject>
	<dc:subject xml:lang="en-US">XGBoost</dc:subject>
	<dc:subject xml:lang="en-US">Long Short-Term Memory</dc:subject>
	<dc:description xml:lang="en-US">Floods remain one of the most destructive and frequent natural hazards, especially in urban river basins like Palembang, Indonesia. Improving early warning systems through the integration of radar and meteorological data has become increasingly feasible with advances in remote sensing and machine learning. The research contribution is the development of a dual-model spatiotemporal prediction framework that combines Sentinel-1 SAR imagery and meteorological data using eXtreme Gradient Boosting (XGBoost) for spatial classification and Long Short-Term Memory (LSTM) for temporal forecasting. Radar backscatter values were extracted from Sentinel-1 Single Look Complex (SLC) data and processed into one-kilometer resolution grids using SNAP and QGIS. These were merged with weather variables precipitation, humidity, wind speed, and solar radiation sourced from local meteorological stations. The XGBoost model achieved a precision of 98.3%, accuracy of 99.94%, recall of 97.5%, and F1-score of 97.9%, with SHAP analysis identifying rainfall and wind speed as dominant flood predictors. Spatial predictions aligned closely with historically flood-prone areas along the Musi River. In contrast, the LSTM model, despite forecasting floods up to 12 days in advance with average accuracy of 91.6%, suffered from class imbalance, resulting in a recall of only 22.9% and precision of 36.3%, limiting its applicability for real-time early warning. These findings demonstrate that while spatial classification using radar and weather data is highly effective, temporal forecasting remains challenged by data imbalance and uneven class distribution. Future research should explore cost-sensitive learning, uncertainty quantification, and real-time validation to enhance the system’s operational reliability.</dc:description>
	<dc:publisher xml:lang="en-US">Universitas Muhammadiyah Yogyakarta</dc:publisher>
	<dc:date>2026-02-14</dc:date>
	<dc:type>info:eu-repo/semantics/article</dc:type>
	<dc:type>info:eu-repo/semantics/publishedVersion</dc:type>
	<dc:format>application/pdf</dc:format>
	<dc:identifier>https://journal.umy.ac.id/index.php/jrc/article/view/27546</dc:identifier>
	<dc:identifier>10.18196/jrc.v6i6.27546</dc:identifier>
	<dc:source xml:lang="en-US">Journal of Robotics and Control (JRC); Vol. 6 No. 6 (2025); 3198-3213</dc:source>
	<dc:source>2715-5072</dc:source>
	<dc:source>2715-5056</dc:source>
	<dc:language>eng</dc:language>
	<dc:relation>https://journal.umy.ac.id/index.php/jrc/article/view/27546/12217</dc:relation>
	<dc:rights xml:lang="en-US">Copyright (c) 2026 Ade Silvia Handayani, Clara Silvia Rotua Aritonang, Suroso Suroso, Nyayu Latifah Husni; Wahyu Caesarendra; Leni Novianti</dc:rights>
	<dc:rights xml:lang="en-US">https://creativecommons.org/licenses/by-sa/4.0</dc:rights>
</oai_dc:dc>