Multischeme ensemble forecasting of surface temperature using neural network over Turkey

Sedef Cakir; Mikdat Kadioglu; Nihat Cubukcu

doi:10.1007/s00704-012-0703-1

Multischeme ensemble forecasting of surface temperature using neural network over Turkey

Sedef Cakir^*, Mikdat Kadioglu, Nihat Cubukcu

^*Corresponding author for this work

Department of Meteorological Engineering

Research output: Contribution to journal › Article › peer-review

9 Citations (Scopus)

Abstract

The ensemble method has long been used to reduce the errors that are caused by initial conditions and/or parameterizations of models in forecasting problems. In this study, neural network (NN) simulations are applied to ensemble weather forecasting. Temperature forecasts averaged over 2 weeks from four different forecasts are used to develop the NN model. Additionally, an ensemble mean of bias-corrected data is used as the control experiment. Overall, ensemble forecasts weighted by NN with feed forward backpropagation algorithm gave better root mean square error, mean absolute error, and same sign percent skills compared to those of the control experiment in most stations and produced more accurate weather forecasts.

Original language	English
Pages (from-to)	703-711
Number of pages	9
Journal	Theoretical and Applied Climatology
Volume	111
Issue number	3-4
DOIs	https://doi.org/10.1007/s00704-012-0703-1
Publication status	Published - Feb 2013

Funding

This research was supported by The Science and Research Council of Turkey (TUBITAK). The FSUGSM data and compiled NCEP Reanalysis were provided by Weather Predict Consulting Inc ( www.weatherpredict.com ). The authors would like to thank Zerefsan Kaymaz, Altug Aksoy, and Samuel Thomas Miller for their valuable review and comments.

Funders	Funder number
TUBITAK
The Science and Research Council of Turkey

Access to Document

10.1007/s00704-012-0703-1

Cite this

@article{ffc8c56be1bf43868b6a61b1f7c2380b,

title = "Multischeme ensemble forecasting of surface temperature using neural network over Turkey",

abstract = "The ensemble method has long been used to reduce the errors that are caused by initial conditions and/or parameterizations of models in forecasting problems. In this study, neural network (NN) simulations are applied to ensemble weather forecasting. Temperature forecasts averaged over 2 weeks from four different forecasts are used to develop the NN model. Additionally, an ensemble mean of bias-corrected data is used as the control experiment. Overall, ensemble forecasts weighted by NN with feed forward backpropagation algorithm gave better root mean square error, mean absolute error, and same sign percent skills compared to those of the control experiment in most stations and produced more accurate weather forecasts.",

author = "Sedef Cakir and Mikdat Kadioglu and Nihat Cubukcu",

year = "2013",

month = feb,

doi = "10.1007/s00704-012-0703-1",

language = "English",

volume = "111",

pages = "703--711",

journal = "Theoretical and Applied Climatology",

issn = "0177-798X",

publisher = "Springer-Verlag Wien",

number = "3-4",

}

TY - JOUR

T1 - Multischeme ensemble forecasting of surface temperature using neural network over Turkey

AU - Cakir, Sedef

AU - Kadioglu, Mikdat

AU - Cubukcu, Nihat

PY - 2013/2

Y1 - 2013/2

N2 - The ensemble method has long been used to reduce the errors that are caused by initial conditions and/or parameterizations of models in forecasting problems. In this study, neural network (NN) simulations are applied to ensemble weather forecasting. Temperature forecasts averaged over 2 weeks from four different forecasts are used to develop the NN model. Additionally, an ensemble mean of bias-corrected data is used as the control experiment. Overall, ensemble forecasts weighted by NN with feed forward backpropagation algorithm gave better root mean square error, mean absolute error, and same sign percent skills compared to those of the control experiment in most stations and produced more accurate weather forecasts.

AB - The ensemble method has long been used to reduce the errors that are caused by initial conditions and/or parameterizations of models in forecasting problems. In this study, neural network (NN) simulations are applied to ensemble weather forecasting. Temperature forecasts averaged over 2 weeks from four different forecasts are used to develop the NN model. Additionally, an ensemble mean of bias-corrected data is used as the control experiment. Overall, ensemble forecasts weighted by NN with feed forward backpropagation algorithm gave better root mean square error, mean absolute error, and same sign percent skills compared to those of the control experiment in most stations and produced more accurate weather forecasts.

UR - http://www.scopus.com/inward/record.url?scp=84873114930&partnerID=8YFLogxK

U2 - 10.1007/s00704-012-0703-1

DO - 10.1007/s00704-012-0703-1

M3 - Article

AN - SCOPUS:84873114930

SN - 0177-798X

VL - 111

SP - 703

EP - 711

JO - Theoretical and Applied Climatology

JF - Theoretical and Applied Climatology

IS - 3-4

ER -

Multischeme ensemble forecasting of surface temperature using neural network over Turkey

Abstract

Funding

Access to Document

Other files and links

Fingerprint

Cite this