Application of the bayesian processor of ensemble to the combination and calibration of ensemble forecasts

Yi Wang, Xiaomei Zhang, Z. Tóth

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Ensemble forecasts are developed to assess and convey uncertainty in weather forecasts. Unfortunately, ensemble prediction systems (EPS) usually underestimate uncertainty and thus are statistically not reliable. In this study, we apply the Bayesian Processor of Ensemble (BPE), which is an extension of the statistical post-processing method of Bayesian Processor of Forecasts (BPF) to calibrate ensemble forecasts. BPE is performed to obtain a posterior function through the combination of a regression-based likelihood function and a climatological prior. The method is applied to 1–10 day lead time EPS forecasts from the NCEP Global Ensemble Forecast System (GEFS) and the Canadian Meteorological Centre (CMC) of 2-m temperature at 24 stations over the continental United States (CONUS). Continuous rank probability score is used to evaluate the performance of posterior probability forecasts. Results show that post-processed ensembles are much better calibrated than the raw ensemble. In addition, merging two ensemble forecasts by incorporating the CMC ensemble mean as another predictor in addition to GEFS ensemble forecasts is shown to provide more skillful and reliable probabilistic forecasts. BPE has a broad potential use in the future given its flexible framework for calibrating and combining ensemble forecast.

Original languageEnglish
Title of host publicationSignal and Information Processing, Networking and Computers - Proceedings of the 5th International Conference on Signal and Information Processing, Networking and Computers ICSINC
EditorsSonglin Sun, Meixia Fu, Lexi Xu
PublisherSpringer Verlag
Pages487-494
Number of pages8
ISBN (Print)9789811371226
DOIs
Publication statusPublished - Jan 1 2019
Event5th International Conference on Signal and Information Processing, Network and Computers, ICSINC 2018 - Yuzhou, China
Duration: Nov 29 2018Dec 1 2018

Publication series

NameLecture Notes in Electrical Engineering
Volume550
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

Conference5th International Conference on Signal and Information Processing, Network and Computers, ICSINC 2018
CountryChina
CityYuzhou
Period11/29/1812/1/18

Fingerprint

Calibration
Merging
Processing
Temperature
Uncertainty

Keywords

  • Ensemble forecasting
  • Statistical post-processing

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Cite this

Wang, Y., Zhang, X., & Tóth, Z. (2019). Application of the bayesian processor of ensemble to the combination and calibration of ensemble forecasts. In S. Sun, M. Fu, & L. Xu (Eds.), Signal and Information Processing, Networking and Computers - Proceedings of the 5th International Conference on Signal and Information Processing, Networking and Computers ICSINC (pp. 487-494). (Lecture Notes in Electrical Engineering; Vol. 550). Springer Verlag. https://doi.org/10.1007/978-981-13-7123-3_57

Application of the bayesian processor of ensemble to the combination and calibration of ensemble forecasts. / Wang, Yi; Zhang, Xiaomei; Tóth, Z.

Signal and Information Processing, Networking and Computers - Proceedings of the 5th International Conference on Signal and Information Processing, Networking and Computers ICSINC. ed. / Songlin Sun; Meixia Fu; Lexi Xu. Springer Verlag, 2019. p. 487-494 (Lecture Notes in Electrical Engineering; Vol. 550).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wang, Y, Zhang, X & Tóth, Z 2019, Application of the bayesian processor of ensemble to the combination and calibration of ensemble forecasts. in S Sun, M Fu & L Xu (eds), Signal and Information Processing, Networking and Computers - Proceedings of the 5th International Conference on Signal and Information Processing, Networking and Computers ICSINC. Lecture Notes in Electrical Engineering, vol. 550, Springer Verlag, pp. 487-494, 5th International Conference on Signal and Information Processing, Network and Computers, ICSINC 2018, Yuzhou, China, 11/29/18. https://doi.org/10.1007/978-981-13-7123-3_57
Wang Y, Zhang X, Tóth Z. Application of the bayesian processor of ensemble to the combination and calibration of ensemble forecasts. In Sun S, Fu M, Xu L, editors, Signal and Information Processing, Networking and Computers - Proceedings of the 5th International Conference on Signal and Information Processing, Networking and Computers ICSINC. Springer Verlag. 2019. p. 487-494. (Lecture Notes in Electrical Engineering). https://doi.org/10.1007/978-981-13-7123-3_57
Wang, Yi ; Zhang, Xiaomei ; Tóth, Z. / Application of the bayesian processor of ensemble to the combination and calibration of ensemble forecasts. Signal and Information Processing, Networking and Computers - Proceedings of the 5th International Conference on Signal and Information Processing, Networking and Computers ICSINC. editor / Songlin Sun ; Meixia Fu ; Lexi Xu. Springer Verlag, 2019. pp. 487-494 (Lecture Notes in Electrical Engineering).
@inproceedings{8d6f3dcd7384442b8e22f5c331031f58,
title = "Application of the bayesian processor of ensemble to the combination and calibration of ensemble forecasts",
abstract = "Ensemble forecasts are developed to assess and convey uncertainty in weather forecasts. Unfortunately, ensemble prediction systems (EPS) usually underestimate uncertainty and thus are statistically not reliable. In this study, we apply the Bayesian Processor of Ensemble (BPE), which is an extension of the statistical post-processing method of Bayesian Processor of Forecasts (BPF) to calibrate ensemble forecasts. BPE is performed to obtain a posterior function through the combination of a regression-based likelihood function and a climatological prior. The method is applied to 1–10 day lead time EPS forecasts from the NCEP Global Ensemble Forecast System (GEFS) and the Canadian Meteorological Centre (CMC) of 2-m temperature at 24 stations over the continental United States (CONUS). Continuous rank probability score is used to evaluate the performance of posterior probability forecasts. Results show that post-processed ensembles are much better calibrated than the raw ensemble. In addition, merging two ensemble forecasts by incorporating the CMC ensemble mean as another predictor in addition to GEFS ensemble forecasts is shown to provide more skillful and reliable probabilistic forecasts. BPE has a broad potential use in the future given its flexible framework for calibrating and combining ensemble forecast.",
keywords = "Ensemble forecasting, Statistical post-processing",
author = "Yi Wang and Xiaomei Zhang and Z. T{\'o}th",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-981-13-7123-3_57",
language = "English",
isbn = "9789811371226",
series = "Lecture Notes in Electrical Engineering",
publisher = "Springer Verlag",
pages = "487--494",
editor = "Songlin Sun and Meixia Fu and Lexi Xu",
booktitle = "Signal and Information Processing, Networking and Computers - Proceedings of the 5th International Conference on Signal and Information Processing, Networking and Computers ICSINC",

}

TY - GEN

T1 - Application of the bayesian processor of ensemble to the combination and calibration of ensemble forecasts

AU - Wang, Yi

AU - Zhang, Xiaomei

AU - Tóth, Z.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Ensemble forecasts are developed to assess and convey uncertainty in weather forecasts. Unfortunately, ensemble prediction systems (EPS) usually underestimate uncertainty and thus are statistically not reliable. In this study, we apply the Bayesian Processor of Ensemble (BPE), which is an extension of the statistical post-processing method of Bayesian Processor of Forecasts (BPF) to calibrate ensemble forecasts. BPE is performed to obtain a posterior function through the combination of a regression-based likelihood function and a climatological prior. The method is applied to 1–10 day lead time EPS forecasts from the NCEP Global Ensemble Forecast System (GEFS) and the Canadian Meteorological Centre (CMC) of 2-m temperature at 24 stations over the continental United States (CONUS). Continuous rank probability score is used to evaluate the performance of posterior probability forecasts. Results show that post-processed ensembles are much better calibrated than the raw ensemble. In addition, merging two ensemble forecasts by incorporating the CMC ensemble mean as another predictor in addition to GEFS ensemble forecasts is shown to provide more skillful and reliable probabilistic forecasts. BPE has a broad potential use in the future given its flexible framework for calibrating and combining ensemble forecast.

AB - Ensemble forecasts are developed to assess and convey uncertainty in weather forecasts. Unfortunately, ensemble prediction systems (EPS) usually underestimate uncertainty and thus are statistically not reliable. In this study, we apply the Bayesian Processor of Ensemble (BPE), which is an extension of the statistical post-processing method of Bayesian Processor of Forecasts (BPF) to calibrate ensemble forecasts. BPE is performed to obtain a posterior function through the combination of a regression-based likelihood function and a climatological prior. The method is applied to 1–10 day lead time EPS forecasts from the NCEP Global Ensemble Forecast System (GEFS) and the Canadian Meteorological Centre (CMC) of 2-m temperature at 24 stations over the continental United States (CONUS). Continuous rank probability score is used to evaluate the performance of posterior probability forecasts. Results show that post-processed ensembles are much better calibrated than the raw ensemble. In addition, merging two ensemble forecasts by incorporating the CMC ensemble mean as another predictor in addition to GEFS ensemble forecasts is shown to provide more skillful and reliable probabilistic forecasts. BPE has a broad potential use in the future given its flexible framework for calibrating and combining ensemble forecast.

KW - Ensemble forecasting

KW - Statistical post-processing

UR - http://www.scopus.com/inward/record.url?scp=85065486190&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065486190&partnerID=8YFLogxK

U2 - 10.1007/978-981-13-7123-3_57

DO - 10.1007/978-981-13-7123-3_57

M3 - Conference contribution

AN - SCOPUS:85065486190

SN - 9789811371226

T3 - Lecture Notes in Electrical Engineering

SP - 487

EP - 494

BT - Signal and Information Processing, Networking and Computers - Proceedings of the 5th International Conference on Signal and Information Processing, Networking and Computers ICSINC

A2 - Sun, Songlin

A2 - Fu, Meixia

A2 - Xu, Lexi

PB - Springer Verlag

ER -