Table 1. Summary of recent precipitation prediction studies over the past 5 years.
| Ref | Year | Region | Methods | C | R | Period | Forecast horizon | Data type | Performance metrics |
|---|---|---|---|---|---|---|---|---|---|
| Lei et al. (2024) | 2024 | China | CNN, LSTM | √ | 2000–2018 | Monthly | Daily rainfall data from NMIC | RMSE, R2, CC, SIG | |
| Hu, Yin & Guo (2024) | 2024 | France | LSTM, RNN | √ | √ | 2016–2018 | Hourly | Radar data | MSE, MAE, SSIM, CSI, HSS |
| Arbabi et al. (2024) | 2024 | Iran | RF, M5, SVR, GPR, KNN | √ | 1951–2021 | Monthly | Rainfall data from meteorological stations | R2, NS, RMSE, MAE | |
| Ebtehaj & Bonakdari (2024) | 2024 | Canada | CNN, LSTM | √ | 1994–2022 | Hourly | Meteorological data from Quebec province | R2, NSE, AIC, PBIAS, NRMSE, RSR | |
| Putra, Rosid & Handoko (2024) | 2024 | Indonesia | XGBoost | √ | 2022 | Hourly | Rain gauges, weather radar, weather satellite data | RMSE | |
| Liao, Lu & Yin (2024) | 2024 | China | ConvLSTM, SmaAT-UNet | √ | 2009–2015 | Hourly | HKO-7 radar data | POD, CSI, FAR | |
| Wang et al. (2024a) | 2024 | China | LSTM, M-P, GAMMA | √ | 2019–2020 | Hourly | Data from Doppler weather radar, meteorological stations, OTT-Parsivel laser raindrop spectrometer | MRE, MAE, RMSE | |
| Liu et al. (2024) | 2024 | China, India |
DFFNet, CNN | √ | 2016–2019 | Hourly | Northern Xinjiang India, FaceDetection epilepsy NATOPS PEMS-SF |
Accuracy, PRE, REC, F1 | |
| Shejule & Pekkat (2024) | 2024 | India | LSTM | √ | 2015–2019 | Hourly | Meteorological data | RMSE, MAPE | |
| Majnooni et al. (2023) | 2023 | USA | RF, XGBoost, SVR, MLP, KNN, LR, AdaBoost, DT | √ | 1983–2020 | Monthly | Rain gauge data | R, R2, MSE, NSE | |
| Gianoglio et al. (2023) | 2023 | Italy | ANN | √ | 2017–2019 | Daily | Smart rainfall system data | REC, Specificity | |
| Saubhagya et al. (2023) | 2023 | Sri Lanka | Spatial Kriging, CNN, SVM, NB, MLP, LSTM, Logistic Regression, RF | √ | 2015–2019 | Daily | Weather data from MDSL | Accuracy, PRE, REC, F1 | |
| Skarlatos et al. (2023) | 2023 | Greece | Seasonal LSTM, Univariate LSTM | √ | 2010–2020 | Yearly | Meteorological data from GAWSN | MSE | |
| Necesito et al. (2023) | 2023 | Philippines | DWT, Univariate LSTM | √ | 2013–2018 | Monthly | Rainfall data from ASIT | NSE, CC, KGE, IA, LMI, MAPE, RMSE, RSR | |
| Baljon & Sharma (2023) | 2023 | Saudi Arabia | Function Fitting ANN | √ | 1982–2011 | Monthly | Rainfall data from metrological department | Accuracy, PRE, REC, F1, specificity | |
| Kumar et al. (2023) | 2023 | India | CatBoost, XGBoost, Lasso, Ridge, LR, LGBM | √ | 1980–2021 | Daily | Rainfall data from WRIS | MAE, RMSE, RMSPE, R2 | |
| Kwon et al. (2024) | 2023 | Korea | ConvLSTM, U-Net | √ | √ | 2017–2021 | Minutely | Radar data | RMSE, MAE, Accuracy, PRE, REC, F1 |
| Liu et al. (2022) | 2022 | China | ConvLSTM | √ | 2015–2020 | Hourly | Rainfall data from NHB | CC, MSE, RMSE, CSI, FAR, POD | |
| Sulaiman et al. (2022) | 2022 | Malaysia | RF, PCA, SVC, SVR, ANN, RVM |
√ | √ | 1998–2007 | Daily | Atmospheric data and rainfall data |
Accuracy, RMSE, NSE |
| Simanjuntak et al. (2022) | 2022 | Indonesia | Multivariate LSTM, RF | √ | √ | 2021 | Minutely | Himawari-8 and GPM IMERG meteorological data | Accuracy, MAE, RMSE |
| Chu et al. (2022) | 2022 | Korea | SVM, RF, XGBoost | √ | 2007 | Hourly | Rainfall and GIS data | RMSE, MAE, RMSLE | |
| Papailiou et al. (2022) | 2022 | Greece | ANN, MLR | √ | 2006–2018 | Daily | Precipitation data from the NOANN network | NSE, R, RMSE | |
| Di Nunno et al. (2022) | 2022 | Bangladesh | M5P, SVR, M5P-SVR, PSO | √ | 1956–2013 | Monthly | Weather data from BMD | MAE, RMSE, RAE, R2 | |
| Salaeh et al. (2022) | 2022 | Thailand | M5, RF, SVR, MLP, LSTM | √ | 2004–2018 | Monthly | Meteorological data from TMD | MAE, RMSE, R, OI | |
| Poornima et al. (2023) | 2022 | India | LSTM | √ | 1901–2017 | Monthly | Rainfall data from open government data | RMSE, loss, learning rate | |
| Anand & Kannan (2022) | 2022 | India | ANN, RF | √ | 2012–2013 | Daily | Smart rainfall system data | PRE, REC, F1 | |
| Choi et al. (2021) | 2021 | Japan | U-Net | √ | √ | 2017–2019 | Hourly | RAIN-F+ rainfall data | MAE, PPMCC, PRE, REC, F1 |
| Shin et al. (2021) | 2021 | USA Korea |
Regression Tree, RF | √ | 1996–2006 2011–2019 |
Minutely | 2DVD radar data | RMSE, MAE, bias, CORR, COE, 1-NE | |
| Yan et al. (2021) | 2021 | China | TabNet, ANN, LSTM, LightGBM | √ | 2012–2016 | Daily | Meteorological data from stations in China | KGE, MAE, RE, RMSE, MAPE |
|
| Bouget et al. (2021) | 2021 | France | U-Net | √ | 2016–2018 | Hourly | MeteoNet rain radar and wind data | F1, bias, TS | |
| Bellido-Jiménez, Gualda & García-Marín (2021) | 2021 | Spain | MLP, SVM, RF, LI | √ | 2000–2021 | Daily | Precipitation data from RIAA | RMSE, MBE, R2 |
|
| Nguyen, Kim & Bae (2021) | 2021 | Korea | MLR, MARS, MLP, RNN, LSTM | √ | 2016–2020 | Minutely | MAPLE radar data | CSI, POD, PEMR, RMSE, R, RFB | |
| Wang et al. (2021) | 2021 | China | WPD-ELM, ARIMA, BPNN | √ | 1958–2016 | Yearly | Precipitation data from Jinsha weather station | RMSE, MAE, R, NSEC | |
| Narejo et al. (2021) | 2021 | Italy | DBN, CNN | √ | 2010–2015 | Hourly | Meteorological data | MSE, RMSE, R | |
| Zhang et al. (2020) | 2020 | China | CNN, BLSTM | √ | 2014–2016 | Daily | Rain gauge data | RMSE, CC, MBE, MAE | |
| Wehbe, Temimi & Adler (2020) | 2020 | Saudi Arabia | GWR, ANN | √ | 2015–2018 | Daily | Ground-based rainfall data from NCM | RMSE, rBIAS, POD, FAR, PCC, NSE | |
| Chhetri et al. (2020) | 2020 | Bhutan | MLP, CNN, LSTM, GRU, BLSTM, BLSTM-GRU | √ | 1997–2017 | Monthly | Rainfall data from NCHM |
RMSE, MSE, PCC, R2 | |
| Wei & Chou (2020) | 2020 | Taiwan | DNN, MLR | √ | 1961–2017 | Hourly | Typhoon events data | RMSE, rRMSE, MAE, rMAE, R2 |