000908090 001__ 908090
000908090 005__ 20230123110626.0
000908090 0247_ $$2doi$$a10.5194/gmd-2021-430
000908090 0247_ $$2Handle$$a2128/31345
000908090 0247_ $$2altmetric$$aaltmetric:124247858
000908090 037__ $$aFZJ-2022-02369
000908090 041__ $$aEnglish
000908090 082__ $$a550
000908090 1001_ $$0P:(DE-Juel1)177767$$aGong, Bing$$b0$$eCorresponding author
000908090 245__ $$aTemperature forecasting by deep learning methods
000908090 260__ $$aKatlenburg-Lindau$$bCopernicus$$c2022
000908090 3367_ $$0PUB:(DE-HGF)25$$2PUB:(DE-HGF)$$aPreprint$$bpreprint$$mpreprint$$s1655814826_31470
000908090 3367_ $$2ORCID$$aWORKING_PAPER
000908090 3367_ $$028$$2EndNote$$aElectronic Article
000908090 3367_ $$2DRIVER$$apreprint
000908090 3367_ $$2BibTeX$$aARTICLE
000908090 3367_ $$2DataCite$$aOutput Types/Working Paper
000908090 520__ $$aNumerical weather prediction (NWP) models solve a system of partial differential equations based on physical laws to forecast the future state of the atmosphere. These models are deployed operationally, but they are computationally very expensive. Recently, the potential of deep neural networks to generate bespoken weather forecasts has been explored in a couple of scientific studies inspired by the success of video frame prediction models in computer vision. In this study, a simple recurrent neural network with convolutional filters, called ConvLSTM, and an advanced generative network, the Stochastic Adversarial Video Prediction (SAVP) model, are applied to create hourly forecasts of the 2 m temperature for the next 12 hours over Europe. We make use of 13 years of data from the ERA5 reanalysis, of which 11 years are utilized for training and one year each is used for validating and testing. We choose the 2 m temperature, total cloud cover and the 850 hPa temperature as predictors and show that both models attain predictive skill by outperforming persistence forecasts. SAVP is superior to ConvLSTM in terms of several evaluation metrics, confirming previous results from computer vision that larger, more complex networks are better suited to learn complex features and to generate better predictions. The 12-hour forecasts of SAVP attain a mean squared error (MSE) of about 2.3 K2, an anomaly correlation coefficient (ACC) larger than 0.85, a Structural Similarity Index (SSIM) of around 0.72, and a gradient ratio (rG) of about 0.82. The ConvLSTM yields a higher MSE (3.6 K2), a smaller ACC (0.80), and SSIM (0.65), but a slightly larger rG (0.84). The superior performance of SAVP in terms of MSE, ACC, and SSIM can be largely attributed to the generator. A sensitivity study shows that a larger weight of the GAN component in the SAVP loss leads to even better preservation of spatial variability at the cost of a somewhat increased MSE (2.5 K2). Including the 850 hPa temperature as an additional predictor enhances the forecast quality and the model also benefits from a larger spatial domain. By contrast, adding the total cloud cover as predictor or reducing the amount of training data to eight years has only small effects. Although the temperature forecasts obtained in this way are still less powerful than contemporary NWP models, this study demonstrates that sophisticated deep neural networks may achieve considerable forecast quality beyond the nowcasting range in a purely data-driven way.
000908090 536__ $$0G:(DE-HGF)POF4-5111$$a5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x0
000908090 536__ $$0G:(EU-Grant)955513$$aMAELSTROM - MAchinE Learning for Scalable meTeoROlogy and cliMate (955513)$$c955513$$fH2020-JTI-EuroHPC-2019-1$$x1
000908090 536__ $$0G:(EU-Grant)787576$$aIntelliAQ - Artificial Intelligence for Air Quality (787576)$$c787576$$fERC-2017-ADG$$x2
000908090 536__ $$0G:(BMBF)01IS18047A$$aVerbundprojekt DeepRain: Effiziente Lokale Niederschlagsvorhersage durch Maschinelles Lernen (01IS18047A)$$c01IS18047A$$x3
000908090 536__ $$0G:(DE-Juel-1)ESDE$$aEarth System Data Exploration (ESDE)$$cESDE$$x4
000908090 588__ $$aDataset connected to CrossRef
000908090 65027 $$0V:(DE-MLZ)SciArea-140$$2V:(DE-HGF)$$aGeosciences$$x0
000908090 7001_ $$0P:(DE-Juel1)180790$$aLangguth, Michael$$b1
000908090 7001_ $$0P:(DE-Juel1)187069$$aJi, Yan$$b2$$ufzj
000908090 7001_ $$0P:(DE-Juel1)166264$$aMozaffari, Amirpasha$$b3$$ufzj
000908090 7001_ $$0P:(DE-Juel1)180752$$aStadtler, Scarlet$$b4$$ufzj
000908090 7001_ $$0P:(DE-Juel1)187076$$aMache, Karim$$b5$$ufzj
000908090 7001_ $$0P:(DE-Juel1)6952$$aSchultz, Martin G.$$b6
000908090 773__ $$0PERI:(DE-600)2456725-5$$a10.5194/gmd-2021-430$$tGeoscientific model development$$x1991-959X$$y2022
000908090 8564_ $$uhttps://juser.fz-juelich.de/record/908090/files/gmd-2021-430.pdf$$yOpenAccess
000908090 909CO $$ooai:juser.fz-juelich.de:908090$$pdnbdelivery$$pec_fundedresources$$pVDB$$pdriver$$popen_access$$popenaire
000908090 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)177767$$aForschungszentrum Jülich$$b0$$kFZJ
000908090 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)180790$$aForschungszentrum Jülich$$b1$$kFZJ
000908090 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)187069$$aForschungszentrum Jülich$$b2$$kFZJ
000908090 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)166264$$aForschungszentrum Jülich$$b3$$kFZJ
000908090 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)180752$$aForschungszentrum Jülich$$b4$$kFZJ
000908090 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)187076$$aForschungszentrum Jülich$$b5$$kFZJ
000908090 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)6952$$aForschungszentrum Jülich$$b6$$kFZJ
000908090 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5111$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x0
000908090 9141_ $$y2022
000908090 915__ $$0StatID:(DE-HGF)0160$$2StatID$$aDBCoverage$$bEssential Science Indicators$$d2021-01-26
000908090 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
000908090 915__ $$0StatID:(DE-HGF)0113$$2StatID$$aWoS$$bScience Citation Index Expanded$$d2021-01-26
000908090 915__ $$0StatID:(DE-HGF)0700$$2StatID$$aFees$$d2021-01-26
000908090 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000908090 915__ $$0StatID:(DE-HGF)0561$$2StatID$$aArticle Processing Charges$$d2021-01-26
000908090 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2022-11-25
000908090 915__ $$0StatID:(DE-HGF)0501$$2StatID$$aDBCoverage$$bDOAJ Seal$$d2021-01-16T18:00:10Z
000908090 915__ $$0StatID:(DE-HGF)0500$$2StatID$$aDBCoverage$$bDOAJ$$d2021-01-16T18:00:10Z
000908090 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bDOAJ : Peer review$$d2021-01-16T18:00:10Z
000908090 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2022-11-25
000908090 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2022-11-25
000908090 915__ $$0StatID:(DE-HGF)1150$$2StatID$$aDBCoverage$$bCurrent Contents - Physical, Chemical and Earth Sciences$$d2022-11-25
000908090 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bGEOSCI MODEL DEV : 2021$$d2022-11-25
000908090 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2022-11-25
000908090 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search$$d2022-11-25
000908090 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC$$d2022-11-25
000908090 915__ $$0StatID:(DE-HGF)9905$$2StatID$$aIF >= 5$$bGEOSCI MODEL DEV : 2021$$d2022-11-25
000908090 920__ $$lyes
000908090 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0
000908090 980__ $$apreprint
000908090 980__ $$aVDB
000908090 980__ $$aUNRESTRICTED
000908090 980__ $$aI:(DE-Juel1)JSC-20090406
000908090 9801_ $$aFullTexts