Deep Learning Augmented Data Assimilation: Reconstructing Missing Information with Convolutional Autoencoders

Remote sensing data play a critical role in improving numerical weather prediction (NWP). However, the physical principles of radiation dictate that data voids frequently exist in physical space (e.g., subcloud area for satellite infrared radiance or no-precipitation region for radar reflectivity)....

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Monthly weather review Ročník 150; číslo 8; s. 1977 - 1991
Hlavní autoři: Wang, Yueya, Shi, Xiaoming, Lei, Lili, Fung, Jimmy Chi-Hung
Médium: Journal Article
Jazyk:angličtina
Vydáno: Washington American Meteorological Society 01.08.2022
Témata:
Accuracy
Analysis
Approximation
Barotropic mode
Data assimilation
Data augmentation
Data collection
Deep learning
Density
Error reduction
Experiments
Fields
Initial conditions
Machine learning
Mathematical analysis
Numerical prediction
Numerical weather forecasting
Physical states
Precipitation
Probability distribution
Radar
Radar data
Radar reflectivity
Radiance
Radiation
Rain
Reconstruction
Reflectance
Remote observing
Remote sensing
Resolution
Simulation
Voids
Vorticity
Vorticity equations
Weather forecasting
ISSN:0027-0644, 1520-0493
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Remote sensing data play a critical role in improving numerical weather prediction (NWP). However, the physical principles of radiation dictate that data voids frequently exist in physical space (e.g., subcloud area for satellite infrared radiance or no-precipitation region for radar reflectivity). Such data gaps impair the accuracy of initial conditions derived from data assimilation (DA), which has a negative impact on NWP. We use the barotropic vorticity equation to demonstrate the potential of deep learning augmented data assimilation (DDA), which involves reconstructing spatially complete pseudo-observation fields from incomplete observations and using them for DA. By training a convolutional autoencoder (CAE) with a long simulation at a coarse “forecast” resolution (T63), we obtained a deep learning approximation of the “reconstruction operator,” which maps spatially incomplete observations to a model state with full spatial coverage and resolution. The CAE was applied to an incomplete streamfunction observation (∼30% missing) from a high-resolution benchmark simulation and demonstrated satisfactory reconstruction performance, even when only very sparse (1/16 of T63 grid density) observations were used as input. When only spatially incomplete observations are used, the analysis fields obtained from ensemble square root filter (EnSRF) assimilation exhibit significant error. However, in DDA, when EnSRF takes in the combined data from the incomplete observations and CAE reconstruction, analysis error reduces significantly. Such gains are more pronounced with sparse observation and small ensemble size because the DDA analysis is much less sensitive to observation density and ensemble size than the conventional DA analysis, which is based solely on incomplete observations.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0027-0644
1520-0493
DOI:10.1175/MWR-D-21-0288.1