A Deep Learning Method for Bias Correction of ECMWF 24–240 h Forecasts
نویسندگان
چکیده
Abstract Correcting the forecast bias of numerical weather prediction models is important for severe warnings. The refined grid requires direct correction on gridded products, as opposed to correcting data only at individual stations. In this study, a deep learning method called CU-net proposed correct forecasts four variables from European Centre Medium-Range Weather Forecast Integrated Forecasting System global model (ECMWF-IFS): 2-m temperature, relative humidity, 10-m wind speed, and direction, with lead time 24 h 240 in North China. First, problem transformed into an image-to-image translation under architecture, which based convolutional neural networks. Second, ECMWF-IFS ECMWF reanalysis (ERA5) 2005 2018 are used training, validation, testing datasets. predictors labels (ground truth) created using ERA5, respectively. Finally, performance compared conventional method, anomaly observations (ANO). Results show that have lower root mean square error, bias, absolute higher correlation coefficient than those ANO all times h. improves upon terms above evaluation metrics, whereas temperature humidity. For direction forecast, often difficult achieve, also performance.
منابع مشابه
Evaluation of bias-correction methods for streamflow forecasts
Evaluation of bias-correction methods for ensemble streamflow volume forecasts T. Hashino, A. A. Bradley, and S. S. Schwartz University of Wisconsin, Department of Atmospheric and Ocean Sciences, Madison, Wisconsin, USA The University of Iowa, IIHR – Hydroscience & Engineering, Iowa City, Iowa, USA Center for Urban Environmental Research and Education, UMBC, Baltimore, Maryland, USA Received: 1...
متن کاملEvaluation of bias-correction methods for ensemble streamflow volume forecasts
Ensemble prediction systems are used operationally to make probabilistic streamflow forecasts for seasonal time scales. However, hydrological models used for ensemble streamflow prediction often have simulation biases that degrade forecast quality and limit the operational usefulness of the forecasts. This study evaluates three biascorrection methods for ensemble streamflow volume forecasts. Al...
متن کاملA comparative study of quantitative mapping methods for bias correction of ERA5 reanalysis precipitation data
This study evaluates the ability of different quantitative mapping (QM) methods as a bias correction technique for ERA5 reanalysis precipitation data. Climate type and geographical location can affect the performance of the bias correction method due to differences in precipitation characteristics. For this purpose, ERA5 reanalysis precipitation data for the years 1989-2019 for 10 selected syno...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Advances in Atmospheric Sciences
سال: 2021
ISSN: ['0256-1530', '1861-9533']
DOI: https://doi.org/10.1007/s00376-021-0215-y