Integrated Watershed Management

Integrated Watershed Management

Comparison of the Facebook's Prophet, Thornthwaite, and Blaney-Criddle Models for Daily Evapotranspiration Time Series Forecasting (Case Study: Aleshtar County)

Document Type : Original Article

Authors
Hafez Mirzapour
Ali Haghizadeh
Shokufeh Hassanvand
Leila Ghasemi
Department of Range and Watershed Management Engineering, Faculty of Natural Resources, Lorestan University, Khorramabad, Iran
10.22034/iwm.2024.2037072.1168
Abstract
Extended Abstract
Introduction: Evapotranspiration is one of the most important components of the hydrological cycle. However, due to its complexity, it is difficult to estimate as it can be influenced by numerous factors. Estimating evapotranspiration is crucial for studies such as water resources management and global climate change. In this research, due to the high importance of evapotranspiration estimation, modeling and forecasting of evapotranspiration values in the city of Aleshtar and the selection of the most suitable model were addressed.
Materials and methods: In this study, evapotranspiration was simulated using the Blaney-Criddle and Thornthwaite methods, as well as Facebook's Prophet model. Facebook's Prophet model is available in both R and Python programming languages. In the Facebook's Prophet model, the evaporation trend is determined on a weekly, seasonal, and annual basis. For this purpose, the data used in this study were obtained from the Aleshtar weather station for the statistical period of 2017-2023. Initially, the trend of evapotranspiration was investigated using the Mann-Kendall test. Subsequently, the Blaney-Criddle, Thornthwaite, and Facebook's Prophet models were run using the average temperature. Finally, evaluation criteria were used to assess the performance of the models and determine the most suitable one. These criteria included the root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R²), Nash-Sutcliffe efficiency (NSE), Pearson correlation coefficient (Pearson's r), and Willmott's index of agreement (d).
Results and Discussion: Analysis of the time series of precipitation, mean temperature, and mean relative humidity data in relation to evapotranspiration revealed that despite fluctuations in the time series of these factors, there was an increasing trend in mean temperature and mean precipitation; and the evapotranspiration data exhibited an upward trend, indicating an increase in evapotranspiration over the study period. Model performance evaluation results showed that Facebook's Prophet model performed best with the lowest root mean square error (RMSE=1.33), mean absolute error (MAE=0.79), highest coefficient of determination (R²=0.88), Nash-Sutcliffe efficiency (NSE=0.88), Willmott's index of agreement (d=0.967), and Pearson correlation coefficient (Pearson's r = 0.939) compared to other models. The results showed that the mean actual evapotranspiration, as well as the simulated evapotranspiration using the Blaney-Criddle, Thornthwaite, and Facebook's Prophet model methods during the statistical period were 4.06, 5.28, 5.26, and 4.11 mm, respectively. Facebook's Prophet model provided the closest simulation of evapotranspiration compared to the observed values. The Mann-Kendall test confirmed an increasing trend in the data, suggesting a rise in evapotranspiration over the statistical period. Additionally, it was observed that the time series data for the Blaney-Criddle method and Facebook's Prophet model exhibited a relatively regular trend, while the Thornthwaite method showed an irregular trend in evapotranspiration data over time. According to the Facebook's Prophet model, evapotranspiration reached its peak during the dry months, from early July to early October, and was at its lowest during the cold months of January, February, and March. Additionally, a weekly analysis revealed that the highest and lowest evaporation and transpiration rates occurred on Tuesday and Fridays, respectively. Based on the research findings, the performance of the Blaney-Criddle and Thornthwaite methods is very similar.
Conclusion: Based on the results, it can be concluded that due to the suitable accuracy of the Facebook's Prophet model in predicting evapotranspiration, this model can be used in future studies as well. Additionally, based on the results, the one-year forecast trend indicated that the increase in evapotranspiration in Aleshtar County will continue. Therefore, careful planning is necessary to mitigate evapotranspiration. The findings of this research can also be applied to optimize water management strategies.
Keywords
Mann-Kendall test
Evapotranspiration
time series
Facebook's Prophet model

Subjects

Abdelraouf, R. E., El-Shawadfy, M. A., Bakry, A. B., Abdelaal, H. K., El-Shirbeny, M. A., Ragab, R., & Belopukhov, S. L. (2024). Estimating ETO and scheduling crop irrigation using Blaney–Criddle equation when only air-temperature data are available and solving the issue of missing meteorological data in Egypt. In BIO Web of Conferences (Vol. 82, p. 02020). EDP Sciences.  https://doi.org/10.1051/bioconf/20248202020
Adesogan, S. O., & Sasanya, B. F. (2023). Efficiency of indirect and estimated evapotranspiration methods in South Western Nigeria. International Journal of Hydrology Science and Technology, 15(1), 64-77.  https://doi.org/10.1504/IJHST.2021.10041388
Allen, R. G., Bastiaanssen, W., Wright, J. L., Morse, A., Tasumi, M., & Trezza, R. (2007). Evapotranspiration from satellite images for water management and hydrologic balances. Journal of irrigation and drainage engineering. 380-394.  https://doi.org/10.1061/(ASCE)0733-9437(2007)133:4(380)
Amouzegari, P., Panahi, M., Mirnia, S. K., & Daneshi, A. (2020). Estimation of preservation value of groundwater resources from the villagers' perspective in Alashtar Watershed, Iran. Watershed Engineering and Management, 12(1), 57-71. https://doi.org/10.22092/ijwmse.2019.122994.1532  (In Persian)
Arslan, S. (2022). A hybrid forecasting model using LSTM and Prophet for energy consumption with decomposition of time series data. PeerJ Computer Science, 8, e1001. https://doi.org/10.7717/peerj-cs.1001
Ataei, H., Tashakori Hashemi, S. A., & Raveian, M. (2020). Analysis of the trend of evapotranspiration of reference crop at synoptic stations of Khorasan Razavi province. Journal of Climate Research, 1398(38), 113-129. (In Persian)
Basak, A., Rahman, A. S., Das, J., Hosono, T., & Kisi, O. (2022). Drought forecasting using the Prophet model in a semi-arid climate region of western India. Hydrological Sciences Journal, 67(9), 1397-1417.  https://doi.org/10.1080/02626667.2022.2082876
Blaney, H. F. (1952). Determining water requirements in irrigated areas from climatological and irrigation data. Washington Soil Conservation Service, 48.
Brouwer, C., & Heibloem, M. (1986). Irrigation water management: irrigation water needs. Training manual, 3, 1-5.
Cem Kuzucu, F., & Taş, İ. (2024). Comparison of Evapotranspiration Values Calculated with Empirical Methods and ETgage Measurements. environmental and earth sciences . Preprints. Online: 11 April 2024: 1-11. https://doi.org/10.20944/preprints202404.0790.v1
Dastaran, M., Jafari, S., Moslemi, H., Attarchi, S., & Alavipanah, S.K. (2022). Monitoring Bakhtegan wetland using a time series of satellite data on the Google Earth Engine platform and predicting parameters with Facebook’s Prophet model. RS & GIS for Natural Resources. 13(4), 1–20. https://doi.org/10.30495/GIRS.2022.685454  (In Persian)
De Witte, C. (2022). Altering functional connectivity in the brain by learning new associations. BSc-Thesis. Artificial Intelligence. Radboud University.
Ding, L., Yu, Y., & Zhang, S. (2024). Trend Projections of Potential Evapotranspiration in Yangtze River Delta and the Uncertainty. Atmosphere, 15(3), 357.  https://doi.org/10.3390/atmos15030357
Elagib, N. A., Ali, M. M., & Schneider, K. (2024). Evaluation and bias correction of CRU TS4. 05 potential evapotranspiration across vast environments with limited data. Atmospheric Research, 299, 107194.  https://doi.org/10.1016/j.atmosres.2023.107194  
Gharbia, S. S., Smullen, T., Gill, L., Johnston, P., & Pilla, F. (2018). Spatially distributed potential evapotranspiration modeling and climate projections. Science of The Total Environment, 633, 571-592. https://doi.org/10.1016/j.scitotenv.2018.03.208
Granger, R. J. (2000). Satellite-derived estimates of evapotranspiration in the Gediz basin. Journal of Hydrology, 229(1-2), 70-76.  https://doi.org/10.1016/S0022-1694(99)00200-0
Hamed, K. H. (2008). Trend detection in hydrologic data: The Mann–Kendall trend test under the scaling hypothesis. Journal of hydrology, 349(3-4), 350-363. https://doi.org/10.1016/j.jhydrol.2007.11.009
Haris, M. D., Adytia, D., & Ramadhan, A. W. (2022). Air temperature forecasting with long short-term memory and prophet: a case study of Jakarta, Indonesia. In 2022 International Conference on data science and its applications (ICoDSA) (pp. 251-256). IEEE. https://doi.org/10.1109/ICoDSA55874.2022.9862869
Hassanvand, M., Borna, R., Zohurian Pardel, M., & Shakiba, A. (2022). Study and evaluation of temperature in Aleshtar city based on artificial neural network model. Journal of Geography and Environmental Studies, 11(41), 155-170. (In Persian)
Heidari Motlagh, A., Nasrolahi, A., Sharifipour, M., & Veysi, S. (2021). Evaluation of Different Models for Estimating Reference Evapotranspiration (ETo) in Aleshtar Plain. Iranian Journal of Soil and Water Research, 52(5), 1409-1421. https://doi.org/10.22059/ijswr.2021.319027.668894
Heydari, S., Bromand, S., Zeinali, B., & Pourghasemi, E. (2024). Precipitation Forecast of Three Stations of Lorestan Province in the Next 20 Years. Journal of Environmental Science Studies, 9(1), 7963-7976. https://doi.org/10.22034/jes s.2023.394271.2014
Hosseini, S. M., Ganji Khorramdel, N., Kheltabadi Farahani, A. H. (2016). Empirical and intelligence Models Evaluation in Estimation of Reference Evapotranspiration by Minimum Climate Data; case study shahrekord, Irrigation and Water Engineering, 7(1), pp. 128-141.
Le Houérou, H. N. (1996). Climate change, drought and desertification. Journal of arid Environments, 34(2), 133-185. https://doi.org/10.1006/jare.1996.0099
Liu, S., Bai, J., Jia, Z., Jia, L., Zhou, H., & Lu, L. (2010). Estimation of evapotranspiration in the Mu Us Sandland of China. Hydrology and Earth System Sciences, 14(3), 573-584. https://doi.org/10.5194/hess-14-573-2010
Mobasheri, M., Khavarian, H., Ziaian, P., & Kamali, G. (2005). Estimation of real evaporation and transpiration using MODIS images and Sabal algorithm. 84th Geomatic Conference, Tehran. 1-12. (In Persian)
Oo, Z. Z., & Phyu, S. (2020). Time series prediction based on Facebook Prophet: a case study, temperature forecasting in Myintkyina. International Journal of Applied Mathematics Electronics and Computers, 8(4), 263-267. https://doi.org/10.18100/ijamec.816894
Pour Yazdankhah, H., Razavipour, T., Khaledian, M., & Rezaei, M. (2013). Determining the appropriate methods to estimate reference evaporation and transpiration in Rasht region. The third national conference on comprehensive management of water resources, Sari (In Persian)
Rahman, A. S., Hosono, T., Kisi, O., Dennis, B., & Imon, A. R. (2020). A minimalistic approach for evapotranspiration estimation using the Prophet model. Hydrological Sciences Journal, 65(12), 1994-2006. https://doi.org/10.1080/02626667.2020.1787416
Ramezani, M., Islamian, S.S., Aghakhani, A., & Mirzaei, S.M.J. (2015). Selection of the best real evaporation and transpiration equation through lysimetry data. National Congress of Irrigation and Drainage of Iran. 1-8 (In Persian)
Rumsey, C. A., Miller, M. P., Schwarz, G. E., Hirsch, R. M., & Susong, D. D. (2017). The role of baseflow in dissolved solids delivery to streams in the Upper Colorado River Basin. Hydrological Processes, 31(26), 4705-4718. https://doi.org/10.1002/hyp.13647
Salas, J. D. (1993). Analysis and modelling of hydrological time series. Handbook of hydrology, McGraw-Hill, New York, 19.1-19.72.
Sang, Y. F., Wang, Z., & Liu, C. (2014). Comparison of the MK test and EMD method for trend identification in hydrological time series. Journal of Hydrology, 510, 293-298. https://doi.org/10.1016/j.jhydrol.2013.12.039
Sarıgöl, M., & Katipoğlu, O.M., (2024). Estimation of monthly evaporation values using gradient boosting machines and mode decomposition techniques in the Southeast Anatolia Project (GAP) area in Turkey. Acta Geophysica, 72(2), pp.999-1016. https://doi.org/10.1007/s11600-023-01067-8
Satrio, C. B. A., Darmawan, W., Nadia, B. U., & Hanafiah, N. (2021). Time series analysis and forecasting of coronavirus disease in Indonesia using ARIMA model and PROPHET. Procedia Computer Science, 179, 524-532.  https://doi.org/10.1016/j.procs.2021.01.036
Shabani, M., Asadi, M. A., & Fathian, H. (2024). Improving the daily pan evaporation estimation of long short-term memory and support vector regression models by using the Wild Horse Optimizer algorithm. Water Supply, 24(4), 1315-1334. https://doi.org/10.2166/ws.2024.063
Shahedi, K., & Zarei, M. (2011). Assessment of potential evapotranspiration estimation methods in Mazandaran Province. Irrigation and Water Engineering, 1(3), 12-21. (In Persian)
Taylor, S. J., & Letham, B. (2018). Forecasting at scale. The American Statistician, 72(1), 37-45. https://doi.org/10.1080/00031305.2017.1380080
Thiyagarajan, K., Kodagoda, S., Ulapane, N., & Prasad, M. (2020). A temporal forecasting driven approach using facebook’s prophet method for anomaly detection in sewer air temperature sensor system. In 2020 15th IEEE Conference on industrial electronics and applications (ICIEA) (pp. 25-30). IEEE. https://doi.org/10.1109/ICIEA48937.2020.9248142
Vishwas, B. V., & Patel, A. (2020). Hands-on Time Series Analysis with Python. From Basics to Bleeding Edge Techniques. Berkeley, CA: Apress. (xvii, 407 pages). https://doi.org/10.1007/978-1-4842-5992-4
Wang, J., Cao, X., Cui, X., Wang, H., Zhang, H., Wang, K., Li, X., Li, Z., & Zhou, Y. (2024). Recent advances of green electricity generation: potential in solar interfacial evaporation system. Advanced Materials, 36(16), 2311151. https://doi.org/10.1002/adma.202311151
Willmott, C. J., Rowe, C. M., & Mintz, Y. (1985). Climatology of the terrestrial seasonal water cycle. Journal of Climatology, 5(6), 589-606.  https://doi.org/10.1002/joc.3370050602  
Xiao, Q., Zhou, L., Xiang, X., Liu, L., Liu, X., Li, X., & Ao, T. (2022). Integration of hydrological model and time series model for improving the runoff simulation: a case study on BTOP model in Zhou River Basin, China. Applied Sciences, 12(14), 6883.  https://doi.org/10.3390/app12146883

  • Receive Date 29 July 2024
  • Revise Date 15 September 2024
  • Accept Date 12 October 2024