Water resource management and disaster risk reduction depend on accurate rainfall-runoff modeling (RRM). Time-series data frequently exhibits fine details and local variations that are difficult for traditional LSTM models to capture. To overcome these problems, we introduce an improved RRM model that utilizes a spatial attention-enhanced transductive LSTM (TLSTM) network. By using transductive learning on data points that closely resemble the test set, this model improves performance and captures subtle temporal differences. With the fusion of a spatial attention mechanism, the model can focus on the most important parts of the input. It also includes a more sophisticated differential evolution (DE) algorithm to facilitate complex hyperparameter tuning. For the DE algorithm, we used a mutation strategy that finds a significant cluster using K-means clustering. We applied the Catchment Attributes and Meteorology for Large Sample Studies (CAMELS) dataset and used individual and regional RRM throughout our assessment of the model. There were very positive results for individual basins with an 8-day runoff prediction of 0.728 Nash-Sutcliffe efficiency (NSE). For regional assessment, the model had an NSE of 0.878. This method supports that combining TLSTM with spatial attention and sophisticated DE algorithms increases the accuracy and reliability of rainfall-runoff (RR) predictions, providing opportunities for improved planning and disaster management of water resources. The source code is publicly available at https://github.com/ZhaoChenchina/RRM.
LiuHLiuWLuoJ, et al.Comparative study of multi-combination models for medium-and long-term runoff prediction in Weihe River. IEEE Access2023; 11: 97099–97106.
2.
IshidaKErcanANagasatoT, et al.Use of one-dimensional CNN for input data size reduction in LSTM for improved computational efficiency and accuracy in hourly rainfall-runoff modeling. J Environ Manag2024; 359: 120931.
3.
WangWGaoJLiuZ, et al.A hybrid rainfall-runoff model: integrating initial loss and LSTM for improved forecasting. Front Environ Sci2023; 11: 1–13.
4.
AoulmiYMaroufNAmirecheM, et al.Highly accurate prediction model for daily runoff in semi-arid basin exploiting metaheuristic learning algorithms. Ieee Access2021; 9: 92500–92515.
5.
LiKMaJGaoJ, et al.Resilience assessment of urban distribution network under heavy rain: a knowledge-informed data-driven approach. IEEE Access2023; 11: 63741–63750.
6.
XiaoYWenTGuP, et al.Response of floods to the underlying surface changes in the taojiang river basin using the hydrologic engineering center’s hydrologic modeling system. Water (Basel)2024; 16: 1120.
7.
JawaleMPSThubeA. Rainfall-runoff modeling of urban floods using GIS and HEC-HMS. MethodsX2025; 15: 103437.
8.
WuKHuMZhangY, et al.A modified Xinanjiang model for quantifying streamflow components in a typical watershed in eastern China. Hydrology2024; 11: 90.
9.
LiZJianSGuR, et al.Runoff simulation under the effects of the modified soil water assessment tool (SWAT) model in the Jiyun River basin. Water (Basel)2023; 15: 2110.
10.
SharmaAKumarAShankarV, et al.Hydrological modeling for rainfall–runoff assessment in a sub-himalayan catchment using a distributed MIKE SHE/MIKE+ model. Water Prac Technol2024; 19: wpt2024158.
11.
RubinsteinBJ. The effect of time step on HSPF model performance. Illinois Institute of Technology, 2023.
12.
DavidARodriguezERSchmalzB. Importance of catchment hydrological processes and calibration of hydrological-hydrodynamic rainfall-runoff models in small rural catchments. J Flood Risk Manag2023; 16: e12901.
13.
DhankarBSinhaJJaiswalRK. Assessment of Rainfall-Runoff using HEC-HMS in ravishankar sagar reservoir, dhamtari, chhattisgarh, India. Int J Environ Clim Change2025; 15: 1–15.
14.
HrourYThomasZRousseau-GueutinP, et al.Enhancing hydrological modeling with bias-corrected satellite weather data in data-scarce catchments: a comparative analysis of SWAT and GR4J models. Front Water2025; 7: 1582589.
15.
GichamoTNouraniVGökçekuşH, et al.Ensemble rainfall–runoff modeling of physically based semi-distributed models using multi-source rainfall data fusion. J Water Clim Change2024; 15: 325–347.
16.
ZamaniMGNikooMRRastadD, et al.A comparative study of data-driven models for runoff, sediment, and nitrate forecasting. J Environ Manag2023; 341: 118006.
17.
MohseniUMuskulaSB. Rainfall-runoff modeling using artificial neural network—a case study of purna sub-catchment of Upper Tapi Basin, India. Environ Sci Proc2023; 25: 1.
BigdeliZMajnooni-HerisADelirhasanniaR, et al.Application of support vector machine and boosted tree algorithm for rainfall-runoff modeling (case study: tabriz plain). Environ Water Eng2023; 9: 532–547.
20.
SimsekOCitakogluHGumusV. Applying machine learning to understand Rainfall–Runoff interactions in the Tigris River Basin of Turkey: O. Simsek et al. Pure Appl Geophys2025; 182: 1–32.
21.
BelinaYKassaAKMasindeM. Integrating machine learning and physical models for rainfall-runoff prediction in upper Baro Akobo River Basin, Ethiopia. Hydrol Sci J2025; 70.
22.
BigdeliZMajnooni-HerisADelirhasanniaR, et al.Performance evaluation of random forest and boosted tree in rainfall-runoff process modeling for sub-basins of Lake Urmia. Atmósfera2025; 39: 143–167.
23.
MuñozPMuñozDFOrellana-AlvearJ, et al.Enhancing runoff forecasting through the integration of satellite precipitation data and hydrological knowledge into machine learning models. Nat Hazards2025; 121: 3915–3937.
24.
MishraPKDwivediR. Soft computing techniques for Rainfall-Runoff modeling and analysis in river basin. Water Resour Manage2025; 39: 1–23.
25.
BochareRJainAShrivastavaR. Exploring ANFIS hybrid optimization for improved rainfall-runoff predictions: insights from Banjar River Catchment, India. Environ Dev Sustain2025: 1–17.
26.
PanahiMKhosraviKRezaieF, et al.Unveiling global flood hotspots: optimized machine learning techniques for enhanced flood susceptibility modeling. J Hydrol: Regional Stud2025; 58: 102285.
27.
LiWLiuCHuC, et al.Application of a hybrid algorithm of LSTM and transformer based on random search optimization for improving rainfall-runoff simulation. Sci Rep2024; 14: 11184.
28.
LiHZhangCChuW, et al.A process-driven deep learning hydrological model for daily rainfall-runoff simulation. J Hydrol2024; 637: 131434.
29.
PeerigaRRinkuDRBhaskarJU, et al.Real-Time rain prediction in agriculture using AI and IoT: a bi-directional LSTM approach. Eng Technol Appl Sci Res2024; 14: 15805–15812.
30.
HuFYangQYangJ, et al.An adaptive rainfall-runoff model for daily runoff prediction under the changing environment: stream-LSTM. Environ Model Softw2025; 191: 106524.
31.
HuYLiHZhangC, et al.Investigate the rainfall-runoff relationship and hydrological concepts inside LSTM. Environ Model Softw2025; 192: 106527.
32.
ShengXTangYYueS, et al.Characteristics and prediction of reservoir water quality under the Rainfall-Runoff impact by long short-term memory based encoder-decoder model. Water Resour Manage2025; 39: 1281–1299.
33.
JhongY-DChenC-SJhongB-C, et al.Optimization of LSTM parameters for flash flood forecasting using genetic algorithm. Water Resour Manage2024; 38: 1141–1164.
34.
GegenleithnerSPirkerMDorfmannC, et al.Long short-term memory networks for enhancing real-time flood forecasts: a case study for an underperforming hydrologic model. Hydrol Earth Syst Sci2025; 29: 1939–1962.
35.
KowP-YLiouJ-YYangM-T, et al.Advancing climate-resilient flood mitigation: utilizing transformer-LSTM for water level forecasting at pumping stations. Sci Total Environ2024; 927: 172246.
36.
KimC-SKokK-HKimC-R. Quasi-Optimized LSTM approach for river water level forecasting. Water (Basel)2025; 17: 2087.
37.
LuoJZhuDLiD. Classification-enhanced LSTM model for predicting river water levels. J Hydrol2025; 650: 132535.
38.
RatulMJPAkterU. A synergistic approach to water level prediction: integrating nature-inspired optimizing algorithms with deep learning models in the Shibsha River system. J Hydroinform2025; 27(7): jh2025057.
39.
LeščešenITanhapourMPekárováP, et al.Long short-term memory (LSTM) networks for accurate river flow forecasting: a case study on the Morava river basin (Serbia). Water (Basel)2025; 17: 907.
40.
HuYLiHZhangC, et al.Exploring diverse modeling schemes for runoff prediction: an application to 544 basins in China. EGUsphere2025; 2025: 1–51.
41.
HuangNWangHWangX, et al.A non-invasive load identification method considering feature dimensionality reduction and DB-LSTM. Electronics (Basel)2024; 13: 343.
42.
FranciscoRMatosJPMarinheiroR, et al.Application of temporal fusion transformers to run-of-the-river hydropower scheduling. Hydrology2025; 12: 81.
43.
DurandDAguilarJR-MorenoMD. An analysis of the energy consumption forecasting problem in smart buildings using LSTM. Sustainability2022; 14: 13358.
VincentAMJideshP. An improved hyperparameter optimization framework for AutoML systems using evolutionary algorithms. Sci Rep2023; 13: 4737.
46.
BacaninNStoeanCZivkovicM, et al.On the benefits of using metaheuristics in the hyperparameter tuning of deep learning models for energy load forecasting. Energies2023; 16: 1434.
47.
MoravvejSVMousaviradSJOlivaD, et al.An improved de algorithm to optimise the learning process of a bert-based plagiarism detection model. In: 2022 IEEE Congress on Evolutionary Computation (CEC), 2022, pp.1–7: IEEE.
48.
KapoorAPathirajaSMarshallL, et al.DeepGR4J: a deep learning hybridization approach for conceptual rainfall-runoff modelling. Environ Model Softw2023; 169: 105831.
49.
XiaoLZhongMZhaD. Runoff forecasting using machine-learning methods: case study in the middle reaches of Xijiang river. Frontiers in big Data2022; 4: 752406.
50.
SinghAKKumarPAliR, et al.Application of machine learning technique for rainfall-runoff modelling of highly dynamic watersheds. 2022.
51.
AnarakiMVAchiteMFarzinS, et al.Modeling of monthly Rainfall–Runoff using Various machine learning techniques in Wadi Ouahrane Basin, Algeria. Water (Basel)2023; 15: 3576.
52.
ShekarPRMathewAGopiVP. Rainfall-Runoff modelling using SWAT and eight artificial intelligence models in the Murredu Watershed, India. Environ Monit Assess2023; 195: 1041.
53.
FuladipanahMShahhosseiniARathnayakeN, et al.In-depth simulation of rainfall–runoff relationships using machine learning methods. Water Pract Technol2024; 19: wpt2024147.
54.
SezenCŠrajM. Hourly rainfall-runoff modelling by combining the conceptual model with machine learning models in mostly karst Ljubljanica River catchment in Slovenia. Stoch Environ Res Risk Assess2024; 38: 937–961.
55.
ShahGZaidiAQureshiAL, et al.Rainfall-runoff modeling using machine learning in the ungauged urban watershed of Quetta Valley, Balochistan (Pakistan). Earth Sci Inf2024; 17: 1–17.
56.
IamampaiSTalaluxmanaYKanasutJ, et al.Enhancing rainfall–runoff model accuracy with machine learning models by using soil water index to reflect runoff characteristics. Water Sci Technol2024; 89: 368–381.
57.
KanthariaVMehtaDKumarV, et al.Rainfall–runoff modeling using an adaptive neuro-fuzzy inference system considering soil moisture for the Damanganga basin. J Water Clim Change2024; 15: 2518–2531.
58.
HouénafaSEJohnsonORonohEK, et al.Hybridization of stochastic hydrological models and machine learning methods for improving Rainfall-Runoff modelling. Result Eng2025; 25: 104079.
59.
SayamaTYamadaMYamakitaA, et al.Parameter regionalization of large-scale distributed rainfall–runoff models using a conditional probability method. Progress Earth Planetary Sci2025; 12: 17.
60.
KaewthongNKanplumjitTKwanthongN, et al.Machine learning and the GR2 M model for monthly runoff forecasting. Civil Eng J2025; 11: 393–405.
61.
AnhDTThanhDVLeHM, et al.Effect of gradient descent optimizers and dropout technique on deep learning LSTM performance in rainfall-runoff modeling. Water Resour Manage2023; 37: 639–657.
62.
ShekarPRMathewAYeswanthP, et al.A combined deep CNN-RNN network for rainfall-runoff modelling in Bardha Watershed, India. Artif Intell Geosci2024; 5: 100073.
63.
WangSWangWZhaoG. A novel deep learning rainfall–runoff model based on transformer combined with base flow separation. Hydrol Res2024; 55: 576–594.
64.
KimDHanHLeeH, et al.Predicting flood water level using combined hybrid model of Rainfall-Runoff and AI-based models. KSCE J Civil Eng2024; 28: 1580–1593.
65.
YoonSAhnK-H. Self-training approach to improve the predictability of data-driven rainfall-runoff model in hydrological data-sparse regions. J Hydrol2024; 632: 130862.
66.
WangSChenYAhmedM. EWT_Informer: a novel satellite-derived rainfall–runoff model based on informer. J Hydroinform2024; 26: 88–106.
67.
LiCJiaoYKanG, et al.Comparisons of different machine learning-based Rainfall–Runoff simulations under changing environments. Water (Basel)2024; 16: 302.
68.
ShekarPRMathewASharmaKV. A hybrid CNN–RNN model for rainfall–runoff modeling in the potteruvagu watershed of India. CLEAN–Soil Air Water2025; 53: 2300341.
69.
JiangWDangXZhangR. Empowering Regional Rainfall-Runoff Modeling Through Encoder–Decoder Based on Convolutional Neural Networks. Water (20734441)2025; 17.
70.
LiTLinKLanT, et al.LSTM-FKAN coupled with feature extraction technique for precipitation–runoff modeling. J Hydrol2025; 652: 132705.
71.
ZhangCZhouYLuF, et al.Assessing the performance and interpretability of the CNN-LSTM-attention model for daily streamflow forecasting in typical basins of the eastern qinghai-Tibet plateau. Sci Rep2025; 15: 82.
72.
YinHZhaoXZhangX, et al.Multi-step regional rainfall-runoff modeling using pyramidal transformer. J Hydrol2025; 656: 132935.
73.
BaoNLiCYanX, et al.RFM_Trans: runoff forecasting model for catchment flood protection using strategies optimized transformer. Expert Syst Appl2025; 277: 127228.
74.
ZhuSWangZZhangW, et al.Application of the ResNet-transformer model for runoff prediction based on multi-source data fusion. Water Resour Manage2025; 39: 1–20.
75.
XuDmLiZWangWc, et al.Wavetranstimesnet: an enhanced deep learning monthly runoff prediction model based on wavelet transform and transformer architecture. Stoch Environ Res Risk Assess2025; 39: 883–910.
76.
YinHZhengQWeiC, et al.Monthly streamflow forecasting with temporal-periodic transformer. J Hydrol2025; 660: 133308.
77.
JiangHLiDZhaoJ, et al.LMCNN: a lightweight attention-based modern convolutional neural network design for Runoff prediction. IEEE Trans Geosci Remote Sens2025; 63.
78.
AddorNNewmanAJMizukamiN, et al.The CAMELS data set: catchment attributes and meteorology for large-sample studies. Hydrol Earth Syst Sci2017; 21: 5293–5313.
79.
PuriDSihagPThakurMS, et al.Analysis of data splitting on streamflow prediction using random forest. AIMS Environ Sci2024; 11: 593–609.
80.
ZareeihematPMohamadiSValipourJ, et al.Forecasting stock market volatility using housing market indicators: a reinforcement learning-based feature selection approach. IEEE Access2025; 13: 52621–52643.
81.
MousaviradSJSchaeferGKorovinI, et al.An enhanced differential evolution algorithm using a novel clustering-based mutation operator. In: 2021 IEEE international conference on systems, man, and cybernetics (SMC), 2021, pp.176–181: IEEE.
82.
FangZQuSYangX, et al.Rr-Tide: Are Transformers Effective for Runoff Prediction?, Available at SSRN 4896197, vol. 10, 2024.
83.
JiangJChenCLackingerA, et al.MetaTrans-FSTSF: a transformer-based meta-learning framework for few-shot time series forecasting in flood prediction. Remote Sens (Basel)2024; 17: 77.
84.
HeMXuXWuS, et al.Multi-step ahead forecasting of daily streamflow based on the transform-based deep learning model under different scenarios. Sci Rep2025; 15: 5451.
