Bayesian Learning and Relevance Vector Machines Approach for Downscaling of Monthly Precipitation

Okkan, Umut; İnan, Gül

doi:10.1061/(asce)he.1943-5584.0001024

Cited by 30 publications

(13 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…where z i is the scaled normalized value, x i is the data, x min and x max are, respectively, the minimum and maximum values of the data, and x and S are, respectively, the mean and unbiased standard deviation statistics of the data [37].…”

Section: Resultsmentioning

confidence: 99%

“…The kernel functions treated by LSSVM modeling studies are generally some specific functions including linear, spline, polynomial, sigmoid, and Gaussian radial basis [32][33][34][35][36][37]. In previous studies existing in the literature, the Gaussian radial basis function (RBF) was chosen as the kernel function because it can map samples nonlinearly into a higher dimensional space and is able to tackle the situation having nonlinearity [38].…”

Section: Kernel Functionmentioning

confidence: 99%

See 1 more Smart Citation

Least Squares Support Vector Mechanics to Predict the Stability Number of Rubble-Mound Breakwaters

Gedik

2018

Water

View full text Add to dashboard Cite

In coastal engineering, empirical formulas grounded on experimental works regarding the stability of breakwaters have been developed. In recent years, soft computing tools such as artificial neural networks and fuzzy models have started to be employed to diminish the time and cost spent in these mentioned experimental works. To predict the stability number of rubble-mound breakwaters, the least squares version of support vector machines (LSSVM) method is used because it can be assessed as an alternative one to diverse soft computing techniques. The LSSVM models have been operated through the selected seven parameters, which are determined by Mallows’ Cp approach, that are, namely, breakwater permeability, damage level, wave number, slope angle, water depth, significant wave heights in front of the structure, and peak wave period. The performances of the LSSVM models have shown superior accuracy (correlation coefficients (CC) of 0.997) than that of artificial neural networks (ANN), fuzzy logic (FL), and genetic programming (GP), that are all implemented in the related literature. As a result, it is thought that this study will provide a practical way for readers to estimate the stability number of rubble-mound breakwaters with more accuracy.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Kernel Functionmentioning

confidence: 99%

Least Squares Support Vector Mechanics to Predict the Stability Number of Rubble-Mound Breakwaters

Gedik

2018

Water

View full text Add to dashboard Cite

show abstract

“…Projected changes of hydrometeorological variables for different basins in Turkey under SRES of AR4 have been obtained from the simulation of GCMs in version three of the Coupled Model Inter‐comparison Project, CMIP3 (e.g. Okkan and Fistikoglu, ; Okkan and Inan, , ) However, in the present work, projections derived from statistical downscaling techniques are introduced based on CMIP5 data in the hope of providing a reference for investigating the probable climatic change impacts on the precipitation regime in the Basin under different RCPs.…”

Section: Introductionmentioning

confidence: 99%

“…Wilby et al, 1998;Hessami et al, 2008), weather generator (Semenov, 2008) and artificial neural networks can be preferred. Successful downscaling studies carried out with artificial neural networks (ANNs) under different climate change scenarios can be found in the literature (Goyal and Ojha, 2012;Okkan and Fistikoglu, 2014;Okkan, 2015;Okkan and Inan, 2015b).…”

Section: Introductionmentioning

confidence: 99%

Downscaling of monthly precipitation using CMIP5 climate models operated under RCPs

Okkan

Kırdemir

2016

Meteorological Applications

Self Cite

View full text Add to dashboard Cite

Downscaling of general circulation model (GCM) outputs extracted from CMIP5 datasets to monthly precipitation for the Gediz Basin, Turkey, under Representative Concentration Pathways (RCPs) was performed by statistical downscaling models, multi-GCM ensemble and bias correction. The output databases from 12 GCMs were used for the projections. To determine explanatory predictor variables, the correlation analysis was applied between precipitation observed at 39 meteorological stations located over the Basin and potential predictors of ERA-Interim reanalysis data. After setting both artificial neural networks and least-squares support vector machine-based statistical downscaling models calibrated with determined predictor variables, downscaling models producing the most suitable results were chosen for each meteorological station. The selected downscaling model structure for each station was then operated with historical and future scenarios RCP4.5, RCP6.0 and RCP8.5. Afterwards, the monthly precipitation forecasts were obtained from a multi-GCM ensemble based on Bayesian model averaging and bias correction applications. The statistical significance of the foreseen changes for the future period 2015-2050 was investigated using Student's t test. The projected decrease trend in precipitation is significant for the RCP8.5 scenario, whereas it is less significant for the RCP4.5 and RCP6.0 scenarios.

show abstract

“…Çözünürlük bakımından nispeten kaba olan genel dolaşım modelleri (GCM'ler) iklim değişikliğinin yerel ölçekteki hidro-meteorolojik etkilerini değerlendirmede yeterli değildir. Bu nedenle, istasyon ölçeğinde kaba çözünürlüklü atmosferik modellerin etkisini yorumlamak için yüksek çözünürlüklü sonuçlara ihtiyaç duyulmaktadır[21,22,23]. Bu ihtiyaçtan dolayı kaba çözünürlüklü GCM verilerinin ölçek indirgeme yöntemi kullanılarak yerel ölçeğe indirgenmesiyle çalışma alanın iklimsel özelliklerini daha güvenilir bir şekilde temsil eden veri setine ulaşmak mümkündür.…”

unclassified

Farklı yanlılık düzeltme yöntemlerinin istatistiksel ölçeğe indirgenmiş yağış projeksiyonlarına uygulanması

Kırdemir

Okkan

2019

Balıkesir Üniversitesi Fen Bilimleri Enstitüsü Dergisi

Self Cite

View full text Add to dashboard Cite

Statistical downscaling models are very effective tools for downscaling coarseresolution climate models to local scale and are widely used in climate change studies. The different climate models used in the projections of various hydro-meteorological variables affect the performance of the downscaling models due to their inherent bias and can reduce the precision of predictions. Due to this reason, bias correction methods are needed in addition to the downscaling models. In the study prepared, the precipitation projections were obtained by the climate models derived within the framework of different emission scenarios in terms of the 5th Assessment Report of Intergovernmental Panel on Climate Change (IPCC) and the effects of different bias correction methods on precipitation estimations were investigated as well. For this purpose, firstly, the predictor selection which represents the precipitation of Gediz Basin was carried out and then the coarse-resolution climate models were downscaled to station scale by means of the related precipitation predictors. In the study, 12 different global climate models having raw outputs of 2015-2050 future period were utilized and it was aimed to obtain stronger predictions by combining the projections which were derived by these climate models. Subsequent to combination of multi-model projections, the bias existing in predictions were corrected by Quantile Mapping (QM), Equiratio Quantile Mapping (ERQM), Detrended Quantile Mapping (DQM) and Quantile Delta Mapping (QDM), respectively. According to the obtained results including all performance measures, it has been deduced that QM offers the largest error values. On the other side, it has been concluded that QDM method can better reflect relative changes compared to other methods. When performance indices pointing out extreme processes were also investigated, it was observed that QDM was superior in the evaluation of mean-based precipitation projections.

show abstract

Bayesian Learning and Relevance Vector Machines Approach for Downscaling of Monthly Precipitation

Cited by 30 publications

References 46 publications

Least Squares Support Vector Mechanics to Predict the Stability Number of Rubble-Mound Breakwaters

Least Squares Support Vector Mechanics to Predict the Stability Number of Rubble-Mound Breakwaters

Downscaling of monthly precipitation using CMIP5 climate models operated under RCPs

Farklı yanlılık düzeltme yöntemlerinin istatistiksel ölçeğe indirgenmiş yağış projeksiyonlarına uygulanması

Contact Info

Product

Resources

About