Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: 
A Comparative Analysis with 32 Empirical Methods

Bunnel Bernard; Dwayne Shorlon Renville; Linda Francois

doi:doi:10.11648/j.ajwse.20261202.12

Research Article |

| Peer-Reviewed

Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: A Comparative Analysis with 32 Empirical Methods

Bunnel Bernard^*

, Dwayne Shorlon Renville

, Linda Francois

Published in American Journal of Water Science and Engineering (Volume 12, Issue 2)

Received: 25 April 2026 Accepted: 8 May 2026 Published: 18 May 2026

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

This study investigates the capability of an Artificial Neural Network (ANN) model to estimate daily ETo using several meteorological variables in a data-limited region along Guyana’s coast. The ANN was trained on historical data from 2001–2018 and independently evaluated over 2019–2022, with ETo computed by the FAO Penman–Monteith (PM-56) method serving as the reference benchmark. Model performance was assessed using standard statistical indicators, including the root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R²), Nash–Sutcliffe efficiency (NSE), and index of agreement (IoA). In addition, the ANN model was compared against 32 commonly used empirical ETo estimation methods, encompassing temperature-based, radiation-based, and mass transfer-based approaches. Results indicate that the ANN reproduced PM-56 ETo estimates with high accuracy and minimal bias, achieving R² and NSE values exceeding 0.99 across the validation period. The ANN model also consistently outperformed all empirical methods across all performance metrics, demonstrating superior accuracy and robustness. Among conventional methods, the Hargreaves–Samani and Makkink approaches showed comparatively better performance, while mass transfer-based methods exhibited substantial deviations and poorer performance. These findings suggest that ANN-based models can serve as reliable alternatives for daily ETo estimation in regions where complete meteorological inputs for physically based methods are limited, thereby supporting improved water-resource and agricultural decision-making in Guyana and similar environments.

Published in	American Journal of Water Science and Engineering (Volume 12, Issue 2)
DOI	10.11648/j.ajwse.20261202.12
Page(s)	39-63
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2026. Published by Science Publishing Group

Keywords

Reference Evapotranspiration, Artificial Neural Network, Performance Metrics, Georgetown, Penman-Monteith, Model Validation

References

[1]	F. Üneş, B. Taşar, Y. Z. Kaya, and M. Demirci, “The evaluation and comparison of daily reference evapotranspiration with ANN and empirical methods,” International Journal of Engineering and Geosciences. 2018, vol. 3, no. 1, pp. 1–7. https://doi.org/10.26833/ijeg.331332
[2]	R. G. Allen, M. Tasumi, A. Morse, and R. Trezza, “A Landsat-based energy balance and evapotranspiration model in Western US water rights regulation and planning,” Irrigation and Drainage Systems. 2005, vol. 19, no. 3–4, pp. 251–268. https://doi.org/10.1007/s10795-005-5187-z
[3]	F. H. S. Chiew, N. N. Kamaladasa, H. M. Malano, and T. A. McMahon, “Penman-Monteith, FAO-24 reference crop evapotranspiration and Class-A pan data in Australia,” Agricultural Water Management. 1995, vol. 28, no. 1, pp. 9–21. https://doi.org/10.1016/0378-3774(95)01149-E
[4]	J. M. Jacobs, S. Reddy, and J. M. Fitzgerald, Evaluation of reference evapotranspiration methodologies and AFSIRS crop water use simulation model. Final Report. Florida: Florida Department of Environmental Protection, 2001.
[5]	B. Temesgen, S. Eching, B. Davidoff, and K. Frame, “Comparison of some reference evapotranspiration equations for California,” Journal of Irrigation and Drainage Engineering. 2005, vol. 131, no. 1, pp. 73–84. https://doi.org/10.1061/(ASCE)0733-9437(2005)131:1(73)
[6]	J. D. Valiantzas, “Simplified forms for the standardized FAO-56 Penman-Monteith reference evapotranspiration using limited weather data,” Journal of Hydrology. 2013, vol. 505, pp. 13–23. https://doi.org/10.1016/j.jhydrol.2013.09.005
[7]	J. Fan, W. Yue, L. Wu, F. Zhang, H. Cai, X. Wang, X. Lu, and Y. Xiang, “Evaluation of SVM, ELM and four tree-based ensemble models for predicting daily reference evapotranspiration using limited meteorological data in different climates of China,” Agricultural and Forest Meteorology. 2018, vol. 263, pp. 225–241. https://doi.org/10.1016/j.agrformet.2018.08.019
[8]	P. Lobit, A. Gómez Tagle, F. Bautista, and J. P. Lhomme, “Retrieving air humidity, global solar radiation, and reference evapotranspiration from daily temperatures: Development and validation of new methods for Mexico. Part III: Reference evapotranspiration,” Theoretical and Applied Climatology. 2018, vol. 133, no. 3–4, pp. 787–797. https://doi.org/10.1007/s00704-017-2213-7
[9]	M. Kumar, N. S. Raghuwanshi, R. Singh, W. W. Wallender, and W. O. Pruitt, “Estimating evapotranspiration using artificial neural network,” Journal of Irrigation and Drainage Engineering. 2002, vol. 128, no. 4, pp. 224–233. https://doi.org/10.1061/(ASCE)0733-9437(2002)128:4(224)
[10]	S. S. Zanetti, E. F. Sousa, V. P. S. Oliveira, F. T. Almeida, and S. Bernardo, “Estimating evapotranspiration using artificial neural network and minimum climatological data,” https://doi.org/10.1061/ASCE0733-94372007133:283
[11]	F. Parekh, “Estimating reference crop evapotranspiration using neural network fitting,” International Journal of Computer Applications. 2014, vol. 93, no. 7, pp. 12–17. https://doi.org/10.5120/16130-5116
[12]	V. Z. Antonopoulos and A. V. Antonopoulos, “Daily reference evapotranspiration estimates by artificial neural networks technique and empirical equations using limited input climate variables,” Computers and Electronics in Agriculture. 2017, vol. 132, pp. 86–96. https://doi.org/10.1016/j.compag.2016.11.011
[13]	S. O. Eastman and B. Khan, “Statistical analysis of rainfall data: A case study of Georgetown, Guyana,” International Journal of Innovative Research in Multidisciplinary and Pedagogical Studies (IJIRMPS). 2022, vol. 10, no. 6. https://doi.org/10.37082/IJIRMPS.v10.i6.1612
[14]	R. G. Allen, L. S. Pereira, M. Smith. Crop evapotranspiration: Guidelines for computing crop water requirements. Rome: Food and Agriculture Organization of the United Nations; 1998.
[15]	J. S. Bojanowski, sirad: Functions for Calculating Daily Solar Radiation and Evapotranspiration. R package version 2.3-3. Available: http://sirad.r-forge.r-project.org/
[16]	G. H. Hargreaves and Z. A. Samani, “Reference crop evapotranspiration from temperature,” Applied Engineering in Agriculture. 1985, vol. 1, no. 2, pp. 96–99. https://doi.org/10.13031/2013.26773
[17]	S. Trajković, “Hargreaves versus Penman–Monteith under humid conditions,” Journal of Irrigation and Drainage Engineering. 2007, vol. 133, no. 1, pp. 38–42. https://doi.org/10.1061/(ASCE)0733-9437(2007)133:1(38)
[18]	H. Tabari and P. Hosseinzadehtalaei, “Local calibration of the Hargreaves and Priestley–Taylor equations for estimating reference evapotranspiration in arid and cold climates of Iran based on the Penman–Monteith model,” Journal of Hydrologic Engineering. 2011, vol. 16, no. 10, pp. 837–845. https://doi.org/10.1061/(ASCE)HE.1943-5584.0000366
[19]	P. Droogers and R. G. Allen, “Estimating reference evapotranspiration under inaccurate data conditions,” Irrigation and Drainage Systems. 2002, vol. 16, no. 1, pp. 33–45. https://doi.org/10.1023/A:1015508322413
[20]	A. Berti, G. Tardivo, A. Chiaudani, F. Rech, and M. Borin, “Assessing reference evapotranspiration by the Hargreaves method in north-eastern Italy,” Agricultural Water Management. 2014, vol. 140, pp. 20–25. https://doi.org/10.1016/j.agwat.2014.03.015
[21]	U. Dorji, J. E. Olesen, and M. S. Seidenkrantz, “Water balance in the complex mountainous terrain of Bhutan and linkages to land use,” Journal of Hydrology: Regional Studies. 2016, vol. 7, pp. 55–68. https://doi.org/10.1016/j.ejrh.2016.05.001
[22]	W. Baier and G. W. Robertson, “Estimation of latent evaporation from simple weather observations,” Canadian Journal of Plant Science. 1965, vol. 45, no. 3, pp. 276–284. https://doi.org/10.4141/cjps65-051
[23]	M. Ahooghalandari, M. Khiadani, and M. E. Jahromi, “Developing equations for estimating reference evapotranspiration in Australia,” Water Resources Management. 2016, vol. 30, pp. 3815–3828. https://doi.org/10.1007/s11269-016-1386-7
[24]	G. F. Makkink, “Testing the Penman formula by means of lysimeters,” Journal of the Institution of Water Engineers. 1957, vol. 11, pp. 277–288.
[25]	C. H. B. Priestley and R. J. Taylor, “On the assessment of surface heat flux and evaporation using large-scale parameters,” Monthly Weather Review. 1972, vol. 100, no. 2, pp. 81–92. https://doi.org/10.1175/1520-0493(1972)100%3C0081:OTAOSH%3E2.3.CO;2
[26]	M. E. Jensen and H. R. Haise, “Estimating evapotranspiration from solar radiation,” Journal of the Irrigation and Drainage Division. 1963, vol. 89, no. 4, pp. 15–41.
[27]	G. H. Hargreaves, “Moisture availability and crop production,” Transactions of the ASAE. 1975, vol. 18, no. 5, pp. 980–984. https://doi.org/10.13031/2013.36722
[28]	W. Abtew, “Evapotranspiration measurements and modeling for three wetland systems in south Florida,” Journal of the American Water Resources Association. 1996, vol. 32, no. 3, pp. 465–473. https://doi.org/10.1111/j.1752-1688.1996.tb04044.x
[29]	S. Irmak, A. Irmak, R. G. Allen, and J. W. Jones, “Solar and net radiation-based equations to estimate reference evapotranspiration in humid climates,” Journal of Irrigation and Drainage Engineering. 2003, vol. 129, no. 5, pp. 336–347. https://doi.org/10.1061/(ASCE)0733-9437(2003)129:5(336)
[30]	L. Oudin, F. Hervieu, C. Michel, C. Perrin, V. Andréassian, F. Anctil, and C. Loumagne, “Which potential evapotranspiration input for a lumped rainfall–runoff model?” Journal of Hydrology. 2005, vol. 303, no. 1–4, pp. 290–306. https://doi.org/10.1016/j.jhydrol.2004.08.026
[31]	J. Dalton, “Experimental essays, on the constitution of mixed gases; on the force of steam or vapour from water and other liquids in different temperatures, both in a Torricellian vacuum and in air; on evaporation; and on the expansion of gases by heat,” Memoirs of the Literary and Philosophical Society of Manchester. 1798, vol. 5, no. 2, pp. 535–602. Available: https://www.biodiversitylibrary.org/part/308525
[32]	A. Meyer, Über einige Zusammenhänge zwischen Klima und Boden in Europa. Zurich: ETH Zurich. 1926. https://doi.org/10.3929/ethz-a-000092038
[33]	C. Rohwer and C. Rohwer, Evaporation from free water surfaces. https://doi.org/10.22004/AG.ECON.163103
[34]	F. Albrecht, “Die Methoden zur Bestimmung der Verdunstung der natürlichen Erdoberfläche,” Archiv für Meteorologie, Geophysik und Bioklimatologie, Serie B. 1950, vol. 2, no. 12, pp. 1–38.
[35]	World Meteorological Organization Commission for Instruments and Methods of Observation, Measurement and estimation of evaporation and evapotranspiration. Geneva: Secretariat of the World Meteorological Organization, 1966.
[36]	W. Trabert, “Neue Beobachtungen über Verdampfungsgeschwindigkeiten [New observations on evaporation rates],” Meteorologische Zeitschrift. 2007, pp. 261–263.
[37]	B. Brockamp and H. Wenner, “Verdunstungsmessungen auf den Steiner See bei Münster,” Deutsche Gewässerkundliche Mitteilungen. 1963, vol. 7, pp. 149–154.
[38]	W. Mahringer, “Verdunstungsstudien am Neusiedler See,” Archiv für Meteorologie, Geophysik und Bioklimatologie, Serie B. 1970, vol. 18, pp. 1–20. https://doi.org/10.1007/BF02245865
[39]	H. L. Penman, “Natural evaporation from open water, bare soil and grass,” Proceedings of the Royal Society A. 1948, vol. 193, no. 1032, pp. 120–145. https://doi.org/10.1098/rspa.1948.0037
[40]	V. A. Romanenko, “Computation of the autumn soil moisture using a universal relationship for a large area,” Proceedings of the Ukrainian Hydrometeorological Research Institute. 1961, vol. 3, pp. 12–25.
[41]	S. K. Jain, P. C. Nayak, and K. P. Sudheer, “Models for estimating evapotranspiration using artificial neural networks, and their physical interpretation,” Hydrological Processes. 2008, vol. 22, no. 13, pp. 2225–2234. https://doi.org/10.1002/hyp.6819
[42]	K. G. Sheela and S. N. Deepa, “Review on methods to fix number of hidden neurons in neural networks,” Mathematical Problems in Engineering. 2013. https://doi.org/10.1155/2013/425740
[43]	S. Walls, A. D. Binns, J. Levison, and S. MacRitchie, “Prediction of actual evapotranspiration by artificial neural network models using data from a Bowen ratio energy balance station,” Neural Computing and Applications. 2020, vol. 32, no. 17, pp. 14001–14018. https://doi.org/10.1007/s00521-020-04800-2
[44]	G. Hinton, N. Srivastava, and K. Swersky, Neural Networks for Machine Learning: Lecture 6a Overview of Mini-Batch Gradient Descent. https://www.cs.toronto.edu/~tijmen/csc321/slides/lecture_slides_lec6.pdf
[45]	W. B. Alves, G. D. S. Rolim, and L. E. D. O. Aparecido, “Reference evapotranspiration forecasting by artificial neural networks,” Engenharia Agrícola. 2017, vol. 37, no. 6, pp. 1116–1125. https://doi.org/10.1590/1809-4430-eng.agric.v37n6p1116-1125/2017
[46]	J. Maqsood, A. A. Farooque, F. Abbas, T. Esau, X. Wang, B. Acharya, and H. Afzaal, “Application of artificial neural networks to project reference evapotranspiration under climate change scenarios,” Water Resources Management. 2022, vol. 36, no. 3, pp. 835–851. https://doi.org/10.1007/s11269-021-02997-y

Cite This Article

Plain Text BibTeX RIS

APA Style

Bernard, B., Renville, D. S., Francois, L. (2026). Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: A Comparative Analysis with 32 Empirical Methods. American Journal of Water Science and Engineering, 12(2), 39-63. https://doi.org/10.11648/j.ajwse.20261202.12

Copy | Download

ACS Style

Bernard, B.; Renville, D. S.; Francois, L. Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: A Comparative Analysis with 32 Empirical Methods. Am. J. Water Sci. Eng. 2026, 12(2), 39-63. doi: 10.11648/j.ajwse.20261202.12

Copy | Download

AMA Style

Bernard B, Renville DS, Francois L. Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: A Comparative Analysis with 32 Empirical Methods. Am J Water Sci Eng. 2026;12(2):39-63. doi: 10.11648/j.ajwse.20261202.12

Copy | Download

@article{10.11648/j.ajwse.20261202.12,
  author = {Bunnel Bernard and Dwayne Shorlon Renville and Linda Francois},
  title = {Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: 
A Comparative Analysis with 32 Empirical Methods},
  journal = {American Journal of Water Science and Engineering},
  volume = {12},
  number = {2},
  pages = {39-63},
  doi = {10.11648/j.ajwse.20261202.12},
  url = {https://doi.org/10.11648/j.ajwse.20261202.12},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajwse.20261202.12},
  abstract = {This study investigates the capability of an Artificial Neural Network (ANN) model to estimate daily ETo using several meteorological variables in a data-limited region along Guyana’s coast. The ANN was trained on historical data from 2001–2018 and independently evaluated over 2019–2022, with ETo computed by the FAO Penman–Monteith (PM-56) method serving as the reference benchmark. Model performance was assessed using standard statistical indicators, including the root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R2), Nash–Sutcliffe efficiency (NSE), and index of agreement (IoA). In addition, the ANN model was compared against 32 commonly used empirical ETo estimation methods, encompassing temperature-based, radiation-based, and mass transfer-based approaches. Results indicate that the ANN reproduced PM-56 ETo estimates with high accuracy and minimal bias, achieving R2 and NSE values exceeding 0.99 across the validation period. The ANN model also consistently outperformed all empirical methods across all performance metrics, demonstrating superior accuracy and robustness. Among conventional methods, the Hargreaves–Samani and Makkink approaches showed comparatively better performance, while mass transfer-based methods exhibited substantial deviations and poorer performance. These findings suggest that ANN-based models can serve as reliable alternatives for daily ETo estimation in regions where complete meteorological inputs for physically based methods are limited, thereby supporting improved water-resource and agricultural decision-making in Guyana and similar environments.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: 
A Comparative Analysis with 32 Empirical Methods
AU  - Bunnel Bernard
AU  - Dwayne Shorlon Renville
AU  - Linda Francois
Y1  - 2026/05/18
PY  - 2026
N1  - https://doi.org/10.11648/j.ajwse.20261202.12
DO  - 10.11648/j.ajwse.20261202.12
T2  - American Journal of Water Science and Engineering
JF  - American Journal of Water Science and Engineering
JO  - American Journal of Water Science and Engineering
SP  - 39
EP  - 63
PB  - Science Publishing Group
SN  - 2575-1875
UR  - https://doi.org/10.11648/j.ajwse.20261202.12
AB  - This study investigates the capability of an Artificial Neural Network (ANN) model to estimate daily ETo using several meteorological variables in a data-limited region along Guyana’s coast. The ANN was trained on historical data from 2001–2018 and independently evaluated over 2019–2022, with ETo computed by the FAO Penman–Monteith (PM-56) method serving as the reference benchmark. Model performance was assessed using standard statistical indicators, including the root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R2), Nash–Sutcliffe efficiency (NSE), and index of agreement (IoA). In addition, the ANN model was compared against 32 commonly used empirical ETo estimation methods, encompassing temperature-based, radiation-based, and mass transfer-based approaches. Results indicate that the ANN reproduced PM-56 ETo estimates with high accuracy and minimal bias, achieving R2 and NSE values exceeding 0.99 across the validation period. The ANN model also consistently outperformed all empirical methods across all performance metrics, demonstrating superior accuracy and robustness. Among conventional methods, the Hargreaves–Samani and Makkink approaches showed comparatively better performance, while mass transfer-based methods exhibited substantial deviations and poorer performance. These findings suggest that ANN-based models can serve as reliable alternatives for daily ETo estimation in regions where complete meteorological inputs for physically based methods are limited, thereby supporting improved water-resource and agricultural decision-making in Guyana and similar environments.
VL  - 12
IS  - 2
ER  -

Copy | Download

Author Information

Bunnel Bernard

Department of Mathematics, Physics and Statistics, University of Guyana, Georgetown, Guyana

Contact Email

http://orcid.org/0009-0000-0391-6493
Dwayne Shorlon Renville

Department of Mathematics, Physics and Statistics, University of Guyana, Georgetown, Guyana

Contact Email

http://orcid.org/0000-0003-4332-5844
Linda Francois

Department of Mathematics, Physics and Statistics, University of Guyana, Georgetown, Guyana

Contact Email

http://orcid.org/0009-0007-7671-7824

Download PDF

Submit an Article

Sections

Plain Text BibTeX RIS

APA Style

Bernard, B., Renville, D. S., Francois, L. (2026). Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: A Comparative Analysis with 32 Empirical Methods. American Journal of Water Science and Engineering, 12(2), 39-63. https://doi.org/10.11648/j.ajwse.20261202.12

Copy | Download

ACS Style

Bernard, B.; Renville, D. S.; Francois, L. Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: A Comparative Analysis with 32 Empirical Methods. Am. J. Water Sci. Eng. 2026, 12(2), 39-63. doi: 10.11648/j.ajwse.20261202.12

Copy | Download

AMA Style

Bernard B, Renville DS, Francois L. Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: A Comparative Analysis with 32 Empirical Methods. Am J Water Sci Eng. 2026;12(2):39-63. doi: 10.11648/j.ajwse.20261202.12

Copy | Download

@article{10.11648/j.ajwse.20261202.12,
  author = {Bunnel Bernard and Dwayne Shorlon Renville and Linda Francois},
  title = {Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: 
A Comparative Analysis with 32 Empirical Methods},
  journal = {American Journal of Water Science and Engineering},
  volume = {12},
  number = {2},
  pages = {39-63},
  doi = {10.11648/j.ajwse.20261202.12},
  url = {https://doi.org/10.11648/j.ajwse.20261202.12},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajwse.20261202.12},
  abstract = {This study investigates the capability of an Artificial Neural Network (ANN) model to estimate daily ETo using several meteorological variables in a data-limited region along Guyana’s coast. The ANN was trained on historical data from 2001–2018 and independently evaluated over 2019–2022, with ETo computed by the FAO Penman–Monteith (PM-56) method serving as the reference benchmark. Model performance was assessed using standard statistical indicators, including the root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R2), Nash–Sutcliffe efficiency (NSE), and index of agreement (IoA). In addition, the ANN model was compared against 32 commonly used empirical ETo estimation methods, encompassing temperature-based, radiation-based, and mass transfer-based approaches. Results indicate that the ANN reproduced PM-56 ETo estimates with high accuracy and minimal bias, achieving R2 and NSE values exceeding 0.99 across the validation period. The ANN model also consistently outperformed all empirical methods across all performance metrics, demonstrating superior accuracy and robustness. Among conventional methods, the Hargreaves–Samani and Makkink approaches showed comparatively better performance, while mass transfer-based methods exhibited substantial deviations and poorer performance. These findings suggest that ANN-based models can serve as reliable alternatives for daily ETo estimation in regions where complete meteorological inputs for physically based methods are limited, thereby supporting improved water-resource and agricultural decision-making in Guyana and similar environments.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Estimating Reference Evapotranspiration over Georgetown, Guyana, Using an Artificial Neural Network: 
A Comparative Analysis with 32 Empirical Methods
AU  - Bunnel Bernard
AU  - Dwayne Shorlon Renville
AU  - Linda Francois
Y1  - 2026/05/18
PY  - 2026
N1  - https://doi.org/10.11648/j.ajwse.20261202.12
DO  - 10.11648/j.ajwse.20261202.12
T2  - American Journal of Water Science and Engineering
JF  - American Journal of Water Science and Engineering
JO  - American Journal of Water Science and Engineering
SP  - 39
EP  - 63
PB  - Science Publishing Group
SN  - 2575-1875
UR  - https://doi.org/10.11648/j.ajwse.20261202.12
AB  - This study investigates the capability of an Artificial Neural Network (ANN) model to estimate daily ETo using several meteorological variables in a data-limited region along Guyana’s coast. The ANN was trained on historical data from 2001–2018 and independently evaluated over 2019–2022, with ETo computed by the FAO Penman–Monteith (PM-56) method serving as the reference benchmark. Model performance was assessed using standard statistical indicators, including the root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R2), Nash–Sutcliffe efficiency (NSE), and index of agreement (IoA). In addition, the ANN model was compared against 32 commonly used empirical ETo estimation methods, encompassing temperature-based, radiation-based, and mass transfer-based approaches. Results indicate that the ANN reproduced PM-56 ETo estimates with high accuracy and minimal bias, achieving R2 and NSE values exceeding 0.99 across the validation period. The ANN model also consistently outperformed all empirical methods across all performance metrics, demonstrating superior accuracy and robustness. Among conventional methods, the Hargreaves–Samani and Makkink approaches showed comparatively better performance, while mass transfer-based methods exhibited substantial deviations and poorer performance. These findings suggest that ANN-based models can serve as reliable alternatives for daily ETo estimation in regions where complete meteorological inputs for physically based methods are limited, thereby supporting improved water-resource and agricultural decision-making in Guyana and similar environments.
VL  - 12
IS  - 2
ER  -

Copy | Download