Evaluation of soft-computing techniques for pan evaporation estimation


  • AMIT KUMAR Division of Agricultural Engineering, ICAR- Indian Agricultural Research Institute, New Delhi- 110 012, India
  • A. SARANGI ICAR-Indian Institute of Water Management, Bhubaneshwar, Odisha
  • D.K. SINGH Division of Agricultural Engineering, ICAR- Indian Agricultural Research Institute, New Delhi- 110 012, India
  • I. MANI Vasantrao Naik Marathwada Krishi Vidyapeeth, Parvani, Maharashtra- 431 402, India
  • K. K. BANDHYOPADHYAY ICAR- Indian Institute of Water Management, Bhubanesaer- 751 025, India
  • S. DASH Division of Design of Experiments, ICAR-Indian Agricultural Statiscics Research Institute, New Delhi -110 012
  • M. KHANNA Water Technology Center, ICAR- Indian Agricultural Research Institute, New Delhi- 110 012, India




Evaporation, Prediction, Neural network, Irrigation scheduling, LSTM network


Estimation of pan evaporation (Epan)  can be useful in judicious irrigation scheduling for enhancing agricultural water productivity. The aim of  present study was to assess the efficacy of state-of-the-art LSTM and ANN for daily Epan estimation using meteorological data. Besides this, the effect of static time-series (Julian date) as additional input variable was investigated on performance of soft-computing techniques. For this purpose,the models were trained, tested and validated with eight meteorological variables of 37 years by using preceding 1-, 3- and 5- days’ information. Data were partitioned into three groups as training (60%), testing (20%), and validation (20%) components. It was observed that the models performed well (best) with preceding 5-days meteorological information followed by 3-days and 1-day. However, all LSTMs simulated peak value of Epan was more accurate as compared to lower values. Meteorological data with julian date improved the performance of LSTMs (0.75<NSE 1; PBias< 10; KGE 0.75). The ANN trained using only meteorological data (preceding 5-days information) had better performance error statistics among all other ANNs and LSTMs with minimum MAE (0.68 to 0.86),  RMSE (0.93 to 1.22),  PBias (-0.73 to 2.44) and maximum NSE (0.83 to 0.84) and KGE (0.89 to 0.92). Overall, it was inferred that the forecasting of meteorological parameters using a few days preceding information along with Julian date as the time series variables resulted in better estimation of Epan for the study region.


Abed, M., Imteaz, M. A., Ahmed, A. N. and Huang, Y.F. (2022). Modelling monthly pan evaporation utilising Random Forest and deep learning algorithms. Sci. Rep., 12(1): 13132.

Bicalho, K.V., Araujo, L.C., Cui, Y.J. and Dantas, B.T. (2016). Evaluation of empirical methods for estimating potential evaporation values in northeast France. In E3S Web of Conferences, EDP Sciences, 9:16005.

Chowdhury, S., Nanda, M.K., Saha, G. and Deka, N. (2010). Evaluation of different methods for evapotranspiration estimation using automatic weather station data. J. Agrometeorol., 12(1): 85-88. https://doi.org/10.54386/jam.v12i1.1277

Commeh, M.K., Agyei-Agyemang, A., Tawiah, P.O. andAsaaga, B.A. (2022). CFD analysis of a flat bottom institutional cookstove. Sci. Afr., 16: e01117.

Kim, S., Shiri, J., Singh, V.P., Kisi, O. and Landeras, G. (2015). Predicting daily pan evaporation by soft computing models with limited climatic data. Hydrol. Sci. J., 60(6): 1120-1136.

Kingra, P. K., Kaur, P., & Hundal, S. S. (2002). Estimation of PET by various methods and its relationship with mesh covered pan evaporation at Ludhiana. J. Agrometeorol., 4(2): 143–148. https://doi.org/10.54386/jam.v4i2.455

Kumar, A., Deo, M.M., Jeet, P., Kumari, A. and Prakash, O. (2022). Daily rainfall prediction for Bihar using artificial neural networks: Prediction of rainfall using ANN. J. AgriSearch., 9(4): 320-325.

Kumar, A., Sarangi, A., Singh, D.K., Khanna, M. and Singh, M. (2023). Prediction of relative humidity using soft computing techniques. J. Soil Water Conserv., 22(3): 280-286.

Majhi, B., Naidu, D., Mishra, A.P. andSatapathy, S.C. (2020). Improved prediction of daily pan evaporation using Deep-LSTM model. Neural. Comput. Appl., 32:7823-7838. doi.org/10.1007/s00521-019-04127-7

Naresh, R., Kumar, M., Kumar, S., Singh, K. and Sharma, P. (2023). Estimation of reference evapotranspiration using artificial neural network models for semi-arid region of Haryana. J. Agrometeorol., 25(1): 145-150. https://doi.org/10.54386/jam.v25i1.1914

Sharma, V., Singh, P. K., Bhakar, S. R., Yadav, K. K., Lakhawat, S. S. and Singh, M. (2021). Pan evaporation and sensor based approaches of irrigation scheduling for crop water requirement, growth and yield of okra. J. Agrometeorol., 23(4): 389-395. https://doi.org/10.54386/jam.v23i4.142.

Terzi, O. (2013). Daily pan evaporation estimation using gene expression programming and adaptive neural-based fuzzy inference system. Neural Comput. Appl., 23: 1035-1044. doi.org/10.1007/s0052 1-012-1027-x

Thiemig, V., Rojas, R., Zambrano-Bigiarini, M. and De-Roo, A. (2013). Hydrological evaluation of satellite-based rainfall estimates over the Volta and Baro-Akobo Basin. J. Hydrol., 499: 324-338.




How to Cite

KUMAR, A., SARANGI, A., SINGH, D., MANI, I., BANDHYOPADHYAY, K. K., DASH, S., & KHANNA, M. (2024). Evaluation of soft-computing techniques for pan evaporation estimation. Journal of Agrometeorology, 26(1), 56–62. https://doi.org/10.54386/jam.v26i1.2247



Research Paper