Stochastic Parameterization Using Compressed Sensing: Application to the Lorenz-96 Atmospheric Model
Growing set of optimization and regression techniques, based upon sparse representations of signals, to build models from data sets has received widespread attention recently with the advent of compressed sensing. This paper deals with the parameterization of the Lorenz-96 (Lorenz, 1995) model with...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Stockholm University Press
2022-04-01
|
| Series: | Tellus: Series A, Dynamic Meteorology and Oceanography |
| Subjects: | |
| Online Access: | https://a.tellusjournals.se/articles/42 |
| _version_ | 1811194572668141568 |
|---|---|
| author | A. Mukherjee Y. Aydogdu T. Ravichandran N. Sri Namachchivaya |
| author_facet | A. Mukherjee Y. Aydogdu T. Ravichandran N. Sri Namachchivaya |
| author_sort | A. Mukherjee |
| collection | DOAJ |
| description | Growing set of optimization and regression techniques, based upon sparse representations of signals, to build models from data sets has received widespread attention recently with the advent of compressed sensing. This paper deals with the parameterization of the Lorenz-96 (Lorenz, 1995) model with two time-scales that mimics mid-latitude atmospheric dynamics with microscopic convective processes. Compressed sensing is used to build models (vector fields) to emulate the behavior of the fine-scale process, so that explicit simulations become an online benchmark for parameterization. We apply compressed sensing, where the sparse recovery is achieved by constructing a sensing/dictionary matrix from ergodic samples generated by the Lorenz-96 atmospheric model, to parameterize the unresolved variables in terms of resolved variables. Stochastic parameterization is achieved by auto-regressive modelling of noise. We utilize the ensemble Kalman filter for data assimilation, where observations (direct measurements) are assimilated in the low-dimensional stochastic parameterized model to provide predictions. Finally, we compare the predictions of compressed sensing and Wilks’ polynomial regression to demonstrate the potential effectiveness of the proposed methodology. |
| first_indexed | 2024-04-12T00:28:59Z |
| format | Article |
| id | doaj.art-f93ee94e58464d1c816cf6192e9dc044 |
| institution | Directory Open Access Journal |
| issn | 1600-0870 |
| language | English |
| last_indexed | 2024-04-12T00:28:59Z |
| publishDate | 2022-04-01 |
| publisher | Stockholm University Press |
| record_format | Article |
| series | Tellus: Series A, Dynamic Meteorology and Oceanography |
| spelling | doaj.art-f93ee94e58464d1c816cf6192e9dc0442022-12-22T03:55:25ZengStockholm University PressTellus: Series A, Dynamic Meteorology and Oceanography1600-08702022-04-0174110.16993/tellusa.4276Stochastic Parameterization Using Compressed Sensing: Application to the Lorenz-96 Atmospheric ModelA. Mukherjee0Y. Aydogdu1T. Ravichandran2N. Sri Namachchivaya3Department of Applied Mathematics, University of Waterloo, Waterloo, Ontario N2L 3G1Department of Applied Mathematics, University of Waterloo, Waterloo, Ontario N2L 3G1Department of Applied Mathematics, University of Waterloo, Waterloo, Ontario N2L 3G1Department of Applied Mathematics, University of Waterloo, Waterloo, Ontario N2L 3G1Growing set of optimization and regression techniques, based upon sparse representations of signals, to build models from data sets has received widespread attention recently with the advent of compressed sensing. This paper deals with the parameterization of the Lorenz-96 (Lorenz, 1995) model with two time-scales that mimics mid-latitude atmospheric dynamics with microscopic convective processes. Compressed sensing is used to build models (vector fields) to emulate the behavior of the fine-scale process, so that explicit simulations become an online benchmark for parameterization. We apply compressed sensing, where the sparse recovery is achieved by constructing a sensing/dictionary matrix from ergodic samples generated by the Lorenz-96 atmospheric model, to parameterize the unresolved variables in terms of resolved variables. Stochastic parameterization is achieved by auto-regressive modelling of noise. We utilize the ensemble Kalman filter for data assimilation, where observations (direct measurements) are assimilated in the low-dimensional stochastic parameterized model to provide predictions. Finally, we compare the predictions of compressed sensing and Wilks’ polynomial regression to demonstrate the potential effectiveness of the proposed methodology.https://a.tellusjournals.se/articles/42compressed sensingsparse regressionensemble kalman filterauto-regression |
| spellingShingle | A. Mukherjee Y. Aydogdu T. Ravichandran N. Sri Namachchivaya Stochastic Parameterization Using Compressed Sensing: Application to the Lorenz-96 Atmospheric Model Tellus: Series A, Dynamic Meteorology and Oceanography compressed sensing sparse regression ensemble kalman filter auto-regression |
| title | Stochastic Parameterization Using Compressed Sensing: Application to the Lorenz-96 Atmospheric Model |
| title_full | Stochastic Parameterization Using Compressed Sensing: Application to the Lorenz-96 Atmospheric Model |
| title_fullStr | Stochastic Parameterization Using Compressed Sensing: Application to the Lorenz-96 Atmospheric Model |
| title_full_unstemmed | Stochastic Parameterization Using Compressed Sensing: Application to the Lorenz-96 Atmospheric Model |
| title_short | Stochastic Parameterization Using Compressed Sensing: Application to the Lorenz-96 Atmospheric Model |
| title_sort | stochastic parameterization using compressed sensing application to the lorenz 96 atmospheric model |
| topic | compressed sensing sparse regression ensemble kalman filter auto-regression |
| url | https://a.tellusjournals.se/articles/42 |
| work_keys_str_mv | AT amukherjee stochasticparameterizationusingcompressedsensingapplicationtothelorenz96atmosphericmodel AT yaydogdu stochasticparameterizationusingcompressedsensingapplicationtothelorenz96atmosphericmodel AT travichandran stochasticparameterizationusingcompressedsensingapplicationtothelorenz96atmosphericmodel AT nsrinamachchivaya stochasticparameterizationusingcompressedsensingapplicationtothelorenz96atmosphericmodel |