Publication:
Application of Decision Tree Algorithm for Predicting Monthly Pan Evaporation Rate

Simple item page
dc.contributor.authorAbed M.en_US
dc.contributor.authorImteaz M.A.en_US
dc.contributor.authorAhmed A.N.en_US
dc.contributor.authorHuang Y.F.en_US
dc.contributor.authorid36612762700en_US
dc.contributor.authorid6506146119en_US
dc.contributor.authorid58063046300en_US
dc.contributor.authorid55807263900en_US
dc.date.accessioned2023-05-29T09:38:37Z
dc.date.available2023-05-29T09:38:37Z
dc.date.issued2022
dc.descriptionCorrelation methods; Decision trees; Forecasting; Meteorology; Water management; Wind; Decision-tree algorithm; Design models; Evaporation rate; Hydrological models; Input parameter; Irrigation system design; Malaysia; Non-linear phenomenon; Pan evaporation; Waters resources; Evaporationen_US
dc.description.abstractEvaporation is an essential aspect for management of water resources, irrigation system designs, and hydrological modelling. Evaporation is regarded as a complex and nonlinear phenomenon resulting from interactions of multiple climatic factors. This paper presents efficiency of a Decision Tree (DT) machine learning approach in predicting monthly pan evaporation (Ep) through a case study for the Alor Setar region in Malaysia. Daily meteorological data from a weather station in Malaysia was deployed for testing and training the model by utilising weather parameters, including maximum temperature, minimum temperature, solar radiation, relative humidity, and wind speed for the period 2000-2019. Several models were developed by employing various input combinations and other model parameters. To determine the most effective input parameters for the ML model, the Pearson correlation coefficient was used to select the most efficient input parameters (predictors). The developed ML model was compared to Stephens and Stewart, a widely used empirical technique. Model performance was assessed using standard statistical measures. Furthermore, the Taylor diagram was used to assess the accuracy of the investigated model. The findings of the investigations in relation to various performance evaluation criteria show that the proposed DT structure can successfully predict the monthly evaporation rate with a high level of accuracy (R2= 0.946, RMSE=0.220, MAE=0.173, and NSE=0.947). Furthermore, even for the same input sets, the DT model developed in this study outperformed empirical methods and could greatly enhance the accuracy of monthly Ep estimates. � Hydrology and Water Resources Symposium, HWRS 2022. All rights reserved.en_US
dc.description.natureFinalen_US
dc.identifier.epage445
dc.identifier.scopus2-s2.0-85150600589
dc.identifier.spage434
dc.identifier.urihttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85150600589&partnerID=40&md5=b494022b080732379fe1d280a7e50ad3
dc.identifier.urihttps://irepository.uniten.edu.my/handle/123456789/27009
dc.publisherEngineers Australiaen_US
dc.sourceScopus
dc.sourcetitleHydrology and Water Resources Symposium, HWRS 2022
dc.titleApplication of Decision Tree Algorithm for Predicting Monthly Pan Evaporation Rateen_US
dc.typeConference Paperen_US
dspace.entity.typePublication
Files
Collections
SCOPUS