| NSF Org: |
OAC Office of Advanced Cyberinfrastructure (OAC) |
| Recipient: |
|
| Initial Amendment Date: | August 13, 2018 |
| Latest Amendment Date: | August 13, 2018 |
| Award Number: | 1835769 |
| Award Instrument: | Standard Grant |
| Program Manager: |
Alejandro Suarez
alsuarez@nsf.gov (703)292-7092 OAC Office of Advanced Cyberinfrastructure (OAC) CSE Directorate for Computer and Information Science and Engineering |
| Start Date: | October 1, 2018 |
| End Date: | September 30, 2022 (Estimated) |
| Total Intended Award Amount: | $307,426.00 |
| Total Awarded Amount to Date: | $307,426.00 |
| Funds Obligated to Date: |
|
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
615 W 131ST ST NEW YORK NY US 10027-7922 (212)854-6851 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
New York NY US 10027-6902 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): |
Data Cyberinfrastructure, EarthCube |
| Primary Program Source: |
|
| Program Reference Code(s): |
|
| Program Element Code(s): |
|
| Award Agency Code: | 4900 |
| Fund Agency Code: | 4900 |
| Assistance Listing Number(s): | 47.070 |
ABSTRACT
This project targets a difficult problem in weather and climate prediction -- the representation of convection. Accurate representation of convection is important, since a majority of current model predictions depend on it. Unraveling the physics involved in convective conditions, clouds and aerosols may take years of modeling to fully understand; however, a set of machine learning techniques, known as "neural net techniques", may provide enhanced predictability in the interim, and this project explores their potential.
The project develops a Python library enabling the use of machine learning (artificial neural networks) in a broad range of science domains. The focus is on integration of convection and cloud formation within larger-scale climate models, with the Community Earth System Model (CESM) as an initial target. The project develops a new set of machine learning climate model parameterizations to reduce uncertainty in weather and climate predictions. The neural networks will be trained on high-fidelity simulations that explicitly resolve convection. Two types of high-resolution simulations will be used for training the neural networks: 1) an augmented super-parameterized simulation, and 2) a full Global Cloud Resolving Model (GCRM) simulation based on the ICOsahedral Non-hydrostatic (ICON) modelling frameworks provided by the Max Planck Institute, using initial 5km horizontal resolution. The effort has the potential to increase understanding of convection dynamics and processes across scales, and could potentially be implemented to address other scale problems as well, where it is too computationally costly or impractical to represent processes occurring at much finer scales than the main grid resolution.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH
Note:
When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external
site maintained by the publisher. Some full text articles may not yet be available without a
charge during the embargo (administrative interval).
Some links on this page may take you to non-federal websites. Their policies may differ from
this site.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
This project developed a machine learning methodology to emulate detailed simulations at a few kilometer resolution into climate model simulations. The model was able to accurately reproduce the details of convection over both land and ocean regions, including the diurnal cycle. The machine learning emulator was further analyzed using a lower dimensional representation (autoencoder) to understand regimes of convection across the globe. This allowed understanding the major modes of variability of convection and interpreting the complex neural network. The machine learning emulator was successfully coupled to the actual coarse-scale simulation and was numerically stable. Finally, we developed an emulation of complex aerosol aggregates using graph neural networks. Aggregates can be extremely challenging to model and we demonstrated that graphs could understand how small-scale interactions could define the emergent physical properties of the aggregates.
We developed as part of this work a Fortran-Keras bridge that allows connecting Keras machine learning models with Fortran numerical simulation codes that can be used not only in the climate sciences but across computational physical sciences. We further developed algorithms that expand standard neural networks to include strict physical conservation laws such as energy or mass conservation laws. The results from this proposal should have a strong impact on other research or application of machine learning to computational physical sciences.
Last Modified: 02/10/2023
Modified by: Pierre Gentine
Please report errors in award information by writing to: awardsearch@nsf.gov.
