Daily Prediction and Multi-Step Forward Forecasting of Reference Evapotranspiration Using LSTM and Bi-LSTM Models
Precise forecasting of reference evapotranspiration (ET<sub>0</sub>) is one of the critical initial steps in determining crop water requirements, which contributes to the reliable management and long-term planning of the world’s scarce water sources. This study provides daily prediction...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-02-01
|
| Series: | Agronomy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4395/12/3/594 |
| _version_ | 1797473280294125568 |
|---|---|
| author | Dilip Kumar Roy Tapash Kumar Sarkar Sheikh Shamshul Alam Kamar Torsha Goswami Md Abdul Muktadir Hussein M. Al-Ghobari Abed Alataway Ahmed Z. Dewidar Ahmed A. El-Shafei Mohamed A. Mattar |
| author_facet | Dilip Kumar Roy Tapash Kumar Sarkar Sheikh Shamshul Alam Kamar Torsha Goswami Md Abdul Muktadir Hussein M. Al-Ghobari Abed Alataway Ahmed Z. Dewidar Ahmed A. El-Shafei Mohamed A. Mattar |
| author_sort | Dilip Kumar Roy |
| collection | DOAJ |
| description | Precise forecasting of reference evapotranspiration (ET<sub>0</sub>) is one of the critical initial steps in determining crop water requirements, which contributes to the reliable management and long-term planning of the world’s scarce water sources. This study provides daily prediction and multi-step forward forecasting of ET<sub>0</sub> utilizing a long short-term memory network (LSTM) and a bi-directional LSTM (Bi-LSTM) model. For daily predictions, the LSTM model’s accuracy was compared to that of other artificial intelligence-based models commonly used in ET0 forecasting, including support vector regression (SVR), M5 model tree (M5Tree), multivariate adaptive regression spline (MARS), probabilistic linear regression (PLR), adaptive neuro-fuzzy inference system (ANFIS), and Gaussian process regression (GPR). The LSTM model outperformed the other models in a comparison based on Shannon’s entropy-based decision theory, while the M5 tree and PLR models proved to be the lowest performers. Prior to performing a multi-step-ahead forecasting, ANFIS, sequence-to-sequence regression LSTM network (SSR-LSTM), LSTM, and Bi-LSTM approaches were used for one-step-ahead forecasting utilizing the past values of the ET<sub>0</sub> time series. The results showed that the Bi-LSTM model outperformed other models and that the sequence of models in ascending order in terms of accuracies was Bi-LSTM > SSR-LSTM > ANFIS > LSTM. The Bi-LSTM model provided multi-step (5 day)-ahead ET<sub>0</sub> forecasting in the next step. According to the results, the Bi-LSTM provided reasonably accurate and acceptable forecasting of multi-step-forward ET<sub>0</sub> with relatively lower levels of forecasting errors. In the final step, the generalization capability of the proposed best models (LSTM for daily predictions and Bi-LSTM for multi-step-ahead forecasting) was evaluated on new unseen data obtained from a test station, Ishurdi. The model’s performance was assessed on three distinct datasets (the entire dataset and the first and the second halves of the entire dataset) derived from the test dataset between 1 January 2015 and 31 December 2020. The results indicated that the deep learning techniques (LSTM and Bi-LSTM) achieved equally good performances as the training station dataset, for which the models were developed. The research outcomes demonstrated the ability of the developed deep learning models to generalize the prediction capabilities outside the training station. |
| first_indexed | 2024-03-09T20:13:30Z |
| format | Article |
| id | doaj.art-502b0cf5dba3427e9b96203bd5b6008d |
| institution | Directory Open Access Journal |
| issn | 2073-4395 |
| language | English |
| last_indexed | 2024-03-09T20:13:30Z |
| publishDate | 2022-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj.art-502b0cf5dba3427e9b96203bd5b6008d2023-11-24T00:07:26ZengMDPI AGAgronomy2073-43952022-02-0112359410.3390/agronomy12030594Daily Prediction and Multi-Step Forward Forecasting of Reference Evapotranspiration Using LSTM and Bi-LSTM ModelsDilip Kumar Roy0Tapash Kumar Sarkar1Sheikh Shamshul Alam Kamar2Torsha Goswami3Md Abdul Muktadir4Hussein M. Al-Ghobari5Abed Alataway6Ahmed Z. Dewidar7Ahmed A. El-Shafei8Mohamed A. Mattar9Irrigation and Water Management Division, Bangladesh Agricultural Research Institute, Joydebpur, Gazipur 1701, BangladeshGrain Quality and Nutrition Division, Bangladesh Rice Research Institute, Joydebpur, Gazipur 1701, BangladeshIrrigation and Water Management Division, Bangladesh Agricultural Research Institute, Joydebpur, Gazipur 1701, BangladeshDepartment of Veterinary Microbiology, Faculty of Veterinary and Animal Sciences, West Bengal University of Animal and Fishery Sciences, Kolkata 700056, West Bengal, IndiaCentre for Carbon, Water and Food, The University of Sydney, Camperdown, NSW 2570, AustraliaDepartment of Agricultural Engineering, College of Food and Agriculture Sciences, King Saud University, Riyadh 11451, Saudi ArabiaPrince Sultan Bin Abdulaziz International Prize for Water Chair, Prince Sultan Institute for Environmental, Water and Desert Research, King Saud University, Riyadh 11451, Saudi ArabiaDepartment of Agricultural Engineering, College of Food and Agriculture Sciences, King Saud University, Riyadh 11451, Saudi ArabiaDepartment of Agricultural Engineering, College of Food and Agriculture Sciences, King Saud University, Riyadh 11451, Saudi ArabiaDepartment of Agricultural Engineering, College of Food and Agriculture Sciences, King Saud University, Riyadh 11451, Saudi ArabiaPrecise forecasting of reference evapotranspiration (ET<sub>0</sub>) is one of the critical initial steps in determining crop water requirements, which contributes to the reliable management and long-term planning of the world’s scarce water sources. This study provides daily prediction and multi-step forward forecasting of ET<sub>0</sub> utilizing a long short-term memory network (LSTM) and a bi-directional LSTM (Bi-LSTM) model. For daily predictions, the LSTM model’s accuracy was compared to that of other artificial intelligence-based models commonly used in ET0 forecasting, including support vector regression (SVR), M5 model tree (M5Tree), multivariate adaptive regression spline (MARS), probabilistic linear regression (PLR), adaptive neuro-fuzzy inference system (ANFIS), and Gaussian process regression (GPR). The LSTM model outperformed the other models in a comparison based on Shannon’s entropy-based decision theory, while the M5 tree and PLR models proved to be the lowest performers. Prior to performing a multi-step-ahead forecasting, ANFIS, sequence-to-sequence regression LSTM network (SSR-LSTM), LSTM, and Bi-LSTM approaches were used for one-step-ahead forecasting utilizing the past values of the ET<sub>0</sub> time series. The results showed that the Bi-LSTM model outperformed other models and that the sequence of models in ascending order in terms of accuracies was Bi-LSTM > SSR-LSTM > ANFIS > LSTM. The Bi-LSTM model provided multi-step (5 day)-ahead ET<sub>0</sub> forecasting in the next step. According to the results, the Bi-LSTM provided reasonably accurate and acceptable forecasting of multi-step-forward ET<sub>0</sub> with relatively lower levels of forecasting errors. In the final step, the generalization capability of the proposed best models (LSTM for daily predictions and Bi-LSTM for multi-step-ahead forecasting) was evaluated on new unseen data obtained from a test station, Ishurdi. The model’s performance was assessed on three distinct datasets (the entire dataset and the first and the second halves of the entire dataset) derived from the test dataset between 1 January 2015 and 31 December 2020. The results indicated that the deep learning techniques (LSTM and Bi-LSTM) achieved equally good performances as the training station dataset, for which the models were developed. The research outcomes demonstrated the ability of the developed deep learning models to generalize the prediction capabilities outside the training station.https://www.mdpi.com/2073-4395/12/3/594deep learningrecurrent neural networksmachine learning algorithmsreference evapotranspiration |
| spellingShingle | Dilip Kumar Roy Tapash Kumar Sarkar Sheikh Shamshul Alam Kamar Torsha Goswami Md Abdul Muktadir Hussein M. Al-Ghobari Abed Alataway Ahmed Z. Dewidar Ahmed A. El-Shafei Mohamed A. Mattar Daily Prediction and Multi-Step Forward Forecasting of Reference Evapotranspiration Using LSTM and Bi-LSTM Models Agronomy deep learning recurrent neural networks machine learning algorithms reference evapotranspiration |
| title | Daily Prediction and Multi-Step Forward Forecasting of Reference Evapotranspiration Using LSTM and Bi-LSTM Models |
| title_full | Daily Prediction and Multi-Step Forward Forecasting of Reference Evapotranspiration Using LSTM and Bi-LSTM Models |
| title_fullStr | Daily Prediction and Multi-Step Forward Forecasting of Reference Evapotranspiration Using LSTM and Bi-LSTM Models |
| title_full_unstemmed | Daily Prediction and Multi-Step Forward Forecasting of Reference Evapotranspiration Using LSTM and Bi-LSTM Models |
| title_short | Daily Prediction and Multi-Step Forward Forecasting of Reference Evapotranspiration Using LSTM and Bi-LSTM Models |
| title_sort | daily prediction and multi step forward forecasting of reference evapotranspiration using lstm and bi lstm models |
| topic | deep learning recurrent neural networks machine learning algorithms reference evapotranspiration |
| url | https://www.mdpi.com/2073-4395/12/3/594 |
| work_keys_str_mv | AT dilipkumarroy dailypredictionandmultistepforwardforecastingofreferenceevapotranspirationusinglstmandbilstmmodels AT tapashkumarsarkar dailypredictionandmultistepforwardforecastingofreferenceevapotranspirationusinglstmandbilstmmodels AT sheikhshamshulalamkamar dailypredictionandmultistepforwardforecastingofreferenceevapotranspirationusinglstmandbilstmmodels AT torshagoswami dailypredictionandmultistepforwardforecastingofreferenceevapotranspirationusinglstmandbilstmmodels AT mdabdulmuktadir dailypredictionandmultistepforwardforecastingofreferenceevapotranspirationusinglstmandbilstmmodels AT husseinmalghobari dailypredictionandmultistepforwardforecastingofreferenceevapotranspirationusinglstmandbilstmmodels AT abedalataway dailypredictionandmultistepforwardforecastingofreferenceevapotranspirationusinglstmandbilstmmodels AT ahmedzdewidar dailypredictionandmultistepforwardforecastingofreferenceevapotranspirationusinglstmandbilstmmodels AT ahmedaelshafei dailypredictionandmultistepforwardforecastingofreferenceevapotranspirationusinglstmandbilstmmodels AT mohamedamattar dailypredictionandmultistepforwardforecastingofreferenceevapotranspirationusinglstmandbilstmmodels |