INTEGRASI LONG SHORT-TERM MEMORY DAN LARGE LANGUAGE MODEL UNTUK PREDIKSI CURAH HUJAN DAN MITIGASI KEKERINGAN
DOI:
https://doi.org/10.70248/jrsit.v4i1.4455Abstract
Kekeringan merupakan salah satu bencana hidrometeorologi yang berkaitan dengan rendahnya curah hujan dan dapat berdampak pada ketersediaan air, pertanian, serta aktivitas masyarakat. Penelitian ini bertujuan mengembangkan sistem prediksi curah hujan menggunakan Long Short-Term Memory (LSTM) yang diintegrasikan dengan Large Language Model (LLM) melalui Gemini AI untuk mendukung interpretasi kondisi curah hujan dan mitigasi risiko kekeringan di Kabupaten Kendal. Data yang digunakan berupa 447 data historis curah hujan dasarian dari BMKG dengan rentang nilai curah hujan historis sebesar 0–[nilai maksimum] mm. Tahapan penelitian meliputi pembagian data secara kronologis menjadi 80% training, 10% validation, dan 10% testing, normalisasi Min-Max Scaling, pembentukan sliding window 24 periode, pelatihan LSTM, serta iterative forecasting selama 36 periode dasarian. Hasil evaluasi menghasilkan MAE sebesar 20,8357 mm, MSE sebesar 736,7793 mm², RMSE sebesar 27,1437 mm, dan R² sebesar 0,7339. Penentuan potensi risiko kekeringan didasarkan pada kategori curah hujan rendah menggunakan threshold yang ditetapkan dalam penelitian dan tidak menggunakan indeks kekeringan meteorologis seperti SPI atau SPEI. Hasil prediksi kemudian diklasifikasikan berdasarkan threshold tersebut sebelum dikirimkan ke Gemini AI. Gemini AI berperan sebagai komponen interpretasi yang mengeksekusi prompt engineering berbasis hasil klasifikasi untuk menghasilkan informasi dan rekomendasi mitigasi dalam bahasa alami, bukan sebagai penentu utama risiko. Seluruh pipeline prediksi, klasifikasi, dan interpretasi berhasil diimplementasikan dalam aplikasi web berbasis Streamlit. Sistem memiliki keterbatasan berupa potensi akumulasi galat pada iterative forecasting jangka panjang dan belum mempertimbangkan indeks kekeringan meteorologis.
References
Bendi, M. I. (2024). Informasi Peringatan Dini Potensi Kekeringan Meteorologis Provinsi Nusa Tenggara Timur. 7(2), 46–54.
Cohen, I., Littor, N., Elyashar, A., Cohen, O., & Puzis, R. (2026). Progress in Disaster Science Effects of uncertainty and affective content on large language models ’ disaster assessment : A controlled comparison using synthetic disaster data ☆. Progress in Disaster Science, 30(December 2025), 100571. https://doi.org/10.1016/j.pdisas.2026.100571
Desifa Ramdani Minhar, F. A. (2021). MITIGASI BENCANA DALAM MENGATASI KEKERINGAN DI KALURAHAN GAYAMHARJO KAPANEWON PRAMBANAN KABUPATEN SLEMAN DAERAH ISTIMEWA YOGYAKARTA. 5(1), 368–381.
Farman, H., Hussain, M. A., Shaikh, S., & Hassan, S. (2026). Dual framework for rainfall prediction : a multi-seed machine and deep learning evaluation across Pakistan ’ s climatic regimes. 1–48.
Gemma, M. D. A. N. (2025). PREDIKSI CURAH HUJAN DI KABUPATEN BOGOR MENGGUNAKAN LONG SHORT-TERM. 9(1), 147–157.
Hawali, M. L. (2025). Pemilihan Neuron LSTM dan LSTM Bayesian Optimization Untuk Prediksi Curah Hujan Bulanan Berbasis Iklim. 15(2), 376–382.
Hendra, Y., Mukhtar, H., & Hafsari, R. (2023). Jurnal Software Engineering and Information System ( SEIS ) PREDIKSI CURAH HUJAN DI KOTA PEKANBARU MENGGUNAKAN LSTM. 3(2).
Hermawan, T., & Zuliarso, E. (2025). Perbandingan Metode Recurrent Neural Network ( RNN ) dan Long Short-Term Memory ( LSTM ) untuk Prediksi Curah Hujan. 7(2), 1450–1463. https://doi.org/10.47065/bits.v7i2.8099
Hop, F. J., Linneman, R., Schnitzler, B., Bomers, A., & Booij, M. J. (2024). Real time probabilistic inundation forecasts using a LSTM neural network. Journal of Hydrology, 635(June 2023), 131082. https://doi.org/10.1016/j.jhydrol.2024.131082
Hosseini, F., Prieto, C., & Alvarez, C. (2025). Ensemble learning of catchment-wise optimized LSTMs enhances regional rainfall-runoff modelling − case Study : Basque Country , Spain. 646(August 2024). https://doi.org/10.1016/j.jhydrol.2024.132269
Ibrahem, A., Osman, A., Aldahoul, N., Lun, K., Feng, Y., Lin, J., Elshafie, A., Sherif, M., & Najah, A. (2025). Climate Risk Management A review on machine learning models for drought monitoring and forecasting. Climate Risk Management, 50(August), 100758. https://doi.org/10.1016/j.crm.2025.100758
Juliansyah, D., Amlelia, Q., Najiyah, K., Ramdani, R. N., & Nadira, N. (2025). Bencana Kekeringan dalam Perspektif Mitigasi , Rehabilitasi , dan Rekonstruksi serta Dampaknya : Studi Komparatif antara Nusa Tenggara Barat dan Jerman. 4(2), 414–420.
Karimanzira, D., Rauschenbach, T., & Hellmund, T. (2025). Improved Flood Management and Risk Communication Through Large Language Models. 1–25.
Kratzert, F., Gauch, M., Nearing, G., Hochreiter, S., & Klotz, D. (2021). Niederschlags-Abfluss-Modellierung mit Long Short-Term Memory ( LSTM ). 270–280. https://doi.org/10.1007/s00506-021-00767-z
Kratzert, F., Klotz, D., Brenner, C., Schulz, K., & Herrnegger, M. (2018). Rainfall – runoff modelling using Long Short-Term Memory ( LSTM ) networks. 6005–6022.
Kratzert, F., Klotz, D., Herrnegger, M., & Sampson, A. K. (2019). Toward Improved Predictions in Ungauged Basins : Exploiting the Power of Machine Learning. https://doi.org/10.1029/2019WR026065
Mao, G., Wang, M., Liu, J., Wang, Z., Wang, K., Meng, Y., Zhong, R., Wang, H., & Li, Y. (2021). Comprehensive comparison of artificial neural networks and long short-term memory networks for rainfall-runoff simulation. Physics and Chemistry of the Earth, 123, 103026. https://doi.org/10.1016/j.pce.2021.103026
Maulana, M. R., Fazilatunnisa, A., Febriansyah, M. Y., Muiz, A., & Fauzan, I. (2026). Analisis dan Prediksi Curah Hujan Bulanan Kota Serang Berbasis Apache Spark Menggunakan Dataset BPS Provinsi Banten. 2(1), 15–21.
Mceachran, Z., Ghosh, R., Renganathan, A., Sharma, S., Lindsay, K., Steinbach, M., Nieber, J., Duffy, C., & Kumar, V. (2022). Knowledge ‐ Guided Machine Learning for Operational Flood Forecasting. 1–18. https://doi.org/10.1029/2024WR039064
Meteorologi, S., & Inten, R. (2025). Prediksi Curah Hujan Harian Menggunakan Model ARIMA dan LSTM di Stasiun Meteorologi Radin Inten II Lampung. 20(x), 35–44.
Miri, S. M., Kavianpour, M. R., & Alizadeh, M. J. (2026). Delta feature and random forest – enhanced LSTM – attention forecasts with probabilistic postprocessing for rainfall forecasting. 1–17.
Mishra, P., Ray, S., Lal, P., Nair, S. B., Matuka, A., Tashkandy, Y., & Emam, W. (2025). Climate modeling for South Asia : statistical and deep learning for rainfall and temperature prediction. 1–25.
Onyutha, C. (2025). Embracing large language model ( LLM ) technologies in hydrology research OPEN ACCESS Embracing large language model ( LLM ) technologies in hydrology. Llm.
Prasetia, M. R., Dewi, R. C., Asyhar, A. H., & Hamid, A. (2026). Prediksi Curah Hujan Harian Menggunakan Metode Principal Component Analysis ( PCA ) dan Long Short-Term Memory ( LSTM ). 6(1), 1–13.
Press, A. I. N. (2026). npj Natural Hazards Article in Press Rescue plan intelligent generation for natural disasters : an integrated approach based on Large Language Models TI ES IN ES.
Raval, M., Sivashanmugam, P., Pham, V., Gohel, H., & Kaushik, A. (2021). Automated predictive analytics tool for rainfall forecasting. Scientific Reports, 1–13. https://doi.org/10.1038/s41598-021-95735-8
Rifqi, R. M., Hajar, M. R., & Haryanto, W. T. (2025). DETEKSI BANJIR BERBASIS LLM ( LARGE LANGUAGE MODELS ) MENGGUNAKAN DATA TWITTER / X VIA CHATBOT WHATSAPP. 2, 61–67.
Ripai, I., Baswardono, W., Abdurahman, N., Rijanto, E., Afrianto, I., Sumitra, I. D., Informasi, D. S., Indonesia, U. K., Bandung, K., Buatan, K., Networks, G. N., Models, L. L., Banji, M., & Perkotaan, B. (2026). DIGITAL TWINS UNTUK MANAJEMEN BANJIR PERKOTAAN : 10(2), 1972–1978.
Saengtabtim, K., Leelawat, N., Aumnoysombat, R., Saengwongwattana, N., Suktavornprasit, G., Imamura, F., & Tang, J. (2025). Harnessing generative AI for enhanced disaster management : a systematic review. Big Earth Data, 9(4), 651–675. https://doi.org/10.1080/20964471.2025.2521157
Supriatna, D., Anggai, S., Informatika, T., & Pamulang, U. (2025). Analisis Prediksi Curah Hujan di Kota Tangerang Menggunakan Metode LSTM dan GRU. 5(2), 119–131.
Supriyanto, A., Zuliarso, E., Suharmanto, E. T., Amalina, H., & Damaryanti, F. (2024). DROUGHT PREDICTION USING LSTM MODEL WITH STANDARDIZED PRECIPITATION INDEX ON THE NORTH COAST OF CENTRAL JAVA. 5(6), 1873–1882.
Talan, T. (2026). Machine learning-based rainfall prediction across temporal scales : model benchmarking and explainability analysis. 1–17.
Vellore, R. K., Islam, S., Khare, M., & Kulkarni, S. (2026). Scientific Reports Article in Press A deep learning framework for rainfall forecasting in mumbai metropolitan region IN A Deep Learning Framework for Rainfall Forecasting in Mumbai IN. 0–37.
Wang, F. (2025). Deep Learning Model for Real- Time Flood Forecasting in Fast- Flowing Watershed. February 2023, 18–22. https://doi.org/10.1111/jfr3.70036
Wani, O. A., Mahdi, S. S., Kumar, S. S., Gagnon, A. S., Danish, F., Al-ansari, N., & Hendawy, S. El. (2024). Predicting rainfall using machine learning , deep learning , and time series models across an altitudinal gradient in the North-Western Himalayas.
Wiranta, D. O., & Situmorang, A. (2025). Kajian Risiko Bencana Kekeringan di Kabupaten Indramayu. 7(2), 104–128.
Xiang, Z., Yan, J., & Demir, I. (2020). A Rainfall ‐ Runoff Model With LSTM ‐ Based Sequence ‐ to ‐ Sequence Learning Water Resources Research. 1–17. https://doi.org/10.1029/2019WR025326
Xu, F., Ma, J., Li, N., & Cheng, J. C. P. (2025). International Journal of Disaster Risk Reduction Large language model applications in disaster management : An interdisciplinary review. International Journal of Disaster Risk Reduction, 127(March), 105642. https://doi.org/10.1016/j.ijdrr.2025.105642
Zhang, C., Brodeur, Z. P., Steinschneider, S., & Herman, J. D. (2022). Leveraging Spatial Patterns in Precipitation Forecasts Using Deep Learning to Support Regional Water Management Water Resources Research. 1–18. https://doi.org/10.1029/2021WR031910
