Towards a Comprehensive Assessment of Statistical versus Soft Computing Models in Hydrology: Application to Monthly Pan Evaporation Prediction
This paper evaluates six soft computational models along with three statistical data-driven models for the prediction of pan evaporation (EP). Accordingly, improved kriging—as a novel statistical model—is proposed for accurate predictions of EP for two meteorological stations in Turkey. In the stand...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Water |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4441/13/17/2451 |
| _version_ | 1797520786813091840 |
|---|---|
| author | Mohammad Zounemat-Kermani Behrooz Keshtegar Ozgur Kisi Miklas Scholz |
| author_facet | Mohammad Zounemat-Kermani Behrooz Keshtegar Ozgur Kisi Miklas Scholz |
| author_sort | Mohammad Zounemat-Kermani |
| collection | DOAJ |
| description | This paper evaluates six soft computational models along with three statistical data-driven models for the prediction of pan evaporation (EP). Accordingly, improved kriging—as a novel statistical model—is proposed for accurate predictions of EP for two meteorological stations in Turkey. In the standard kriging model, the input data nonlinearity effects are increased by using a nonlinear map and transferring input data from a polynomial to an exponential basic function. The accuracy, precision, and over/under prediction tendencies of the response surface method, kriging, improved kriging, multilayer perceptron neural network using the Levenberg–Marquardt (MLP-LM) as well as a conjugate gradient (MLP-CG), radial basis function neural network (RBFNN), multivariate adaptive regression spline (MARS), M5Tree and support vector regression (SVR) were compared. Overall, all the applied models were highly capable of predicting monthly EP in both stations with a mean absolute error (<i>MAE</i>) < 0.77 mm and a Willmott index (<i>d</i>) > 0.95. Considering periodicity as an input parameter, the MLP-LM provided better results than the other methods among the soft computing models (<i>MAE</i> = 0.492 mm and <i>d</i> = 0.981). However, the improved kriging method surpassed all the other models based on the statistical measures (<i>MAE</i> = 0.471 mm and <i>d</i> = 0.983). Finally, the outcomes of the Mann–Whitney test indicated that the applied soft computational models do not have significant superiority over the statistical ones (<i>p</i>-value > 0.65 at α = 0.01 and α = 0.05). |
| first_indexed | 2024-03-10T08:01:35Z |
| format | Article |
| id | doaj.art-f8eb229db17540908f44230da6f84980 |
| institution | Directory Open Access Journal |
| issn | 2073-4441 |
| language | English |
| last_indexed | 2024-03-10T08:01:35Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Water |
| spelling | doaj.art-f8eb229db17540908f44230da6f849802023-11-22T11:26:03ZengMDPI AGWater2073-44412021-09-011317245110.3390/w13172451Towards a Comprehensive Assessment of Statistical versus Soft Computing Models in Hydrology: Application to Monthly Pan Evaporation PredictionMohammad Zounemat-Kermani0Behrooz Keshtegar1Ozgur Kisi2Miklas Scholz3Department of Water Engineering, Shahid Bahonar University of Kerman, Kerman 7616913439, IranDepartment of Civil Engineering, University of Zabol, Zabol 9861335856, IranDepartment of Civil Engineering, Ilia State University, Tbilisi 0162, GeorgiaDivision of Water Resources Engineering, Faculty of Engineering, Lund University, P.O. Box 118, 22100 Lund, SwedenThis paper evaluates six soft computational models along with three statistical data-driven models for the prediction of pan evaporation (EP). Accordingly, improved kriging—as a novel statistical model—is proposed for accurate predictions of EP for two meteorological stations in Turkey. In the standard kriging model, the input data nonlinearity effects are increased by using a nonlinear map and transferring input data from a polynomial to an exponential basic function. The accuracy, precision, and over/under prediction tendencies of the response surface method, kriging, improved kriging, multilayer perceptron neural network using the Levenberg–Marquardt (MLP-LM) as well as a conjugate gradient (MLP-CG), radial basis function neural network (RBFNN), multivariate adaptive regression spline (MARS), M5Tree and support vector regression (SVR) were compared. Overall, all the applied models were highly capable of predicting monthly EP in both stations with a mean absolute error (<i>MAE</i>) < 0.77 mm and a Willmott index (<i>d</i>) > 0.95. Considering periodicity as an input parameter, the MLP-LM provided better results than the other methods among the soft computing models (<i>MAE</i> = 0.492 mm and <i>d</i> = 0.981). However, the improved kriging method surpassed all the other models based on the statistical measures (<i>MAE</i> = 0.471 mm and <i>d</i> = 0.983). Finally, the outcomes of the Mann–Whitney test indicated that the applied soft computational models do not have significant superiority over the statistical ones (<i>p</i>-value > 0.65 at α = 0.01 and α = 0.05).https://www.mdpi.com/2073-4441/13/17/2451pan evaporationmachine learning modelsimproved krigingSVRMARS |
| spellingShingle | Mohammad Zounemat-Kermani Behrooz Keshtegar Ozgur Kisi Miklas Scholz Towards a Comprehensive Assessment of Statistical versus Soft Computing Models in Hydrology: Application to Monthly Pan Evaporation Prediction Water pan evaporation machine learning models improved kriging SVR MARS |
| title | Towards a Comprehensive Assessment of Statistical versus Soft Computing Models in Hydrology: Application to Monthly Pan Evaporation Prediction |
| title_full | Towards a Comprehensive Assessment of Statistical versus Soft Computing Models in Hydrology: Application to Monthly Pan Evaporation Prediction |
| title_fullStr | Towards a Comprehensive Assessment of Statistical versus Soft Computing Models in Hydrology: Application to Monthly Pan Evaporation Prediction |
| title_full_unstemmed | Towards a Comprehensive Assessment of Statistical versus Soft Computing Models in Hydrology: Application to Monthly Pan Evaporation Prediction |
| title_short | Towards a Comprehensive Assessment of Statistical versus Soft Computing Models in Hydrology: Application to Monthly Pan Evaporation Prediction |
| title_sort | towards a comprehensive assessment of statistical versus soft computing models in hydrology application to monthly pan evaporation prediction |
| topic | pan evaporation machine learning models improved kriging SVR MARS |
| url | https://www.mdpi.com/2073-4441/13/17/2451 |
| work_keys_str_mv | AT mohammadzounematkermani towardsacomprehensiveassessmentofstatisticalversussoftcomputingmodelsinhydrologyapplicationtomonthlypanevaporationprediction AT behroozkeshtegar towardsacomprehensiveassessmentofstatisticalversussoftcomputingmodelsinhydrologyapplicationtomonthlypanevaporationprediction AT ozgurkisi towardsacomprehensiveassessmentofstatisticalversussoftcomputingmodelsinhydrologyapplicationtomonthlypanevaporationprediction AT miklasscholz towardsacomprehensiveassessmentofstatisticalversussoftcomputingmodelsinhydrologyapplicationtomonthlypanevaporationprediction |