Have a personal or library account? Click to login
Evapotranspiration Estimation Using Machine Learning Methods Cover

Evapotranspiration Estimation Using Machine Learning Methods

Journal of Horticultural Research
Volume 31 (2023): Issue 2 (December 2023)
By: Waldemar Treder,  Krzysztof Klamkowski,  Katarzyna Wójcik and  Anna Tryngiel-Gać  
Open Access
|Dec 2023

Figures & Tables

Figure 1.

Relationship between evapotranspiration (ETo) and important weather parameters (Skierniewice 2009–2022)
Relationship between evapotranspiration (ETo) and important weather parameters (Skierniewice 2009–2022)

Figure 2.

The importance of variables when creating regression treesSR – average solar radiation level, VPD – vapor pressure deficit, Tmax – maximum temperature, Tavg – average temperature, RH – relative humidity, Ra – extraterrestrial solar radiation, #D – day number of the year
The importance of variables when creating regression treesSR – average solar radiation level, VPD – vapor pressure deficit, Tmax – maximum temperature, Tavg – average temperature, RH – relative humidity, Ra – extraterrestrial solar radiation, #D – day number of the year

Figure 3.

The importance of variables when creating boosted treesNote: see Figure 2
The importance of variables when creating boosted treesNote: see Figure 2

Figure 4.

The importance of variables when creating random forestsNote: see Figure 2
The importance of variables when creating random forestsNote: see Figure 2

Figure 5.

Net changes in extraterrestrial solar radiation (Ra) and the average level of solar radiation (SR) during the vegetation period (Skierniewice 2009–2022)
Net changes in extraterrestrial solar radiation (Ra) and the average level of solar radiation (SR) during the vegetation period (Skierniewice 2009–2022)

Pearson correlation coefficients between daily evapotranspiration (ETo) and meteorological data

SRTavgTmaxRHU2RaVPD#D
ETo0.940.660.72−0.690.050.620.82−0.41

Statistical analysis of the performance of the RT, BRT, RF, and ANN models in estimating daily ETo with two different meteorological input datasets

ModelRadiationR2SlopeMSERMSE
Regression trees+0.9110.9110.1080.329
−0.8130.8130.2280.478
Boosted trees+0.9420.9310.0730.269
−0.8340.8250.2050.453
Random forests+0.9520.8950.0660.256
−0.8410.7990.2070.455
Artificial neural networks+0.9630.9470.0230.152
−0.8700.8430.0820.286
DOI: https://doi.org/10.2478/johr-2023-0033 | Journal eISSN: 2353-3978 | Journal ISSN: 2300-5009
Journal RSS Feed
Language: English
Page range: 35 - 44
Submitted on: Sep 1, 2023
|
Accepted on: Nov 1, 2023
|
Published on: Dec 29, 2023
Published by: National Institute of Horticultural Research
In partnership with: Paradigm Publishing Services
Publication frequency: 2 issues per year
Keywords:
regression trees,
boosted trees,
random forests,
neural networks,
weather data,
incomplete meteorological data
Related subjects:
Life sciences,
Biotechnology,
Plant science,
Ecology,
Life sciences, other

© 2023 Waldemar Treder, Krzysztof Klamkowski, Katarzyna Wójcik, Anna Tryngiel-Gać, published by National Institute of Horticultural Research
This work is licensed under the Creative Commons Attribution 4.0 License.

Previous articleVolume 31 (2023): Issue 2 (December 2023)Next article