-
Data Imbalance, Uncertainty Quantification, and Generalization via Transfer Learning in Data-driven Parameterizations: Lessons from the Emulation of Gravity Wave Momentum Transport in WACCM
Authors:
Y. Qiang Sun,
Hamid A. Pahlavan,
Ashesh Chattopadhyay,
Pedram Hassanzadeh,
Sandro W. Lubis,
M. Joan Alexander,
Edwin Gerber,
Aditi Sheshadri,
Yifei Guan
Abstract:
Neural networks (NNs) are increasingly used for data-driven subgrid-scale parameterization in weather and climate models. While NNs are powerful tools for learning complex nonlinear relationships from data, there are several challenges in using them for parameterizations. Three of these challenges are 1) data imbalance related to learning rare (often large-amplitude) samples; 2) uncertainty quanti…
▽ More
Neural networks (NNs) are increasingly used for data-driven subgrid-scale parameterization in weather and climate models. While NNs are powerful tools for learning complex nonlinear relationships from data, there are several challenges in using them for parameterizations. Three of these challenges are 1) data imbalance related to learning rare (often large-amplitude) samples; 2) uncertainty quantification (UQ) of the predictions to provide an accuracy indicator; and 3) generalization to other climates, e.g., those with higher radiative forcing. Here, we examine the performance of methods for addressing these challenges using NN-based emulators of the Whole Atmosphere Community Climate Model (WACCM) physics-based gravity wave (GW) parameterizations as the test case. WACCM has complex, state-of-the-art parameterizations for orography-, convection- and frontal-driven GWs. Convection- and orography-driven GWs have significant data imbalance due to the absence of convection or orography in many grid points. We address data imbalance using resampling and/or weighted loss functions, enabling the successful emulation of parameterizations for all three sources. We demonstrate that three UQ methods (Bayesian NNs, variational auto-encoders, and dropouts) provide ensemble spreads that correspond to accuracy during testing, offering criteria on when a NN gives inaccurate predictions. Finally, we show that the accuracy of these NNs decreases for a warmer climate (4XCO2). However, the generalization accuracy is significantly improved by applying transfer learning, e.g., re-training only one layer using ~1% new data from the warmer climate. The findings of this study offer insights for develo** reliable and generalizable data-driven parameterizations for various processes, including (but not limited) to GWs.
△ Less
Submitted 27 November, 2023;
originally announced November 2023.
-
The Influence of Boreal Summer Madden-Julian Oscillation on Precipitation Extremes in Indonesia
Authors:
Fadhil R. Muhammad,
Sandro W. Lubis,
Sonni Setiawan
Abstract:
This study examines the influence of Madden-Julian Oscillation (MJO) on Indonesian precipitation during the extended boreal summer (May to September). The MJO is one of the dominant intraseasonal variabilities that influence the extreme precipitation in the tropics, especially in Indonesia. The episodes of intense precipitation (95th percentiles) during active MJO phases from 1998-2015 are evaluat…
▽ More
This study examines the influence of Madden-Julian Oscillation (MJO) on Indonesian precipitation during the extended boreal summer (May to September). The MJO is one of the dominant intraseasonal variabilities that influence the extreme precipitation in the tropics, especially in Indonesia. The episodes of intense precipitation (95th percentiles) during active MJO phases from 1998-2015 are evaluated using the daily precipitation datasets from the gridded Asian Precipitation Highly Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE) and several station rain gauges. The boreal summer MJO influences the probability of extreme precipitation, especially in the west and north parts of Indonesia. The west part experiences an increase in the probability of extreme precipitation by up to 55\% and 80\% during phases 2 and 3, respectively. Moreover, the extreme precipitation probability in the north part increases by up to 40 to 70\% during phases 2 to 4. On the other hand, the influence of MJO is relatively weak in the south and east parts of Indonesia. The contrast of the precipitation response between the north and south parts of Indonesia is consistent with the northward movement of boreal summer MJO.
△ Less
Submitted 13 October, 2021;
originally announced December 2021.
-
Impacts of the Madden Julian Oscillation on Precipitation Extremes in Indonesia
Authors:
Fadhlil R. Muhammad,
Sandro W. Lubis,
Sonni Setiawan
Abstract:
The influence of the Madden Julian Oscillation (MJO) on the precipitation extremes in Indonesia during the rainy season (October to April) has been evaluated using the daily station rain gauge data and the gridded Asian Precipitation Highly Resolved Observational Data Integration Toward Evaluation of Water Resources (APHRODITE) from 1987 to 2017 for different phases of the MJO. The results show th…
▽ More
The influence of the Madden Julian Oscillation (MJO) on the precipitation extremes in Indonesia during the rainy season (October to April) has been evaluated using the daily station rain gauge data and the gridded Asian Precipitation Highly Resolved Observational Data Integration Toward Evaluation of Water Resources (APHRODITE) from 1987 to 2017 for different phases of the MJO. The results show that MJO significantly modulates the frequency of extreme precipitation events in Indonesia, with the magnitude of the impact varying across regions. Specifically, the convectively active (suppressed) MJO increases (decreases) the probability of extreme precipitation events over the western and central parts of Indonesia by up to 70% (40%). In the eastern part of Indonesia, MJO increases (decreases) extreme precipitation probability by up to 50% (40%). We attribute the differences in the probability of extreme precipitation events to the changes in the horizontal moisture flux convergence induced by MJO. The results indicate that the MJO provides the source of predictability of daily extreme precipitation in Indonesia.
△ Less
Submitted 20 July, 2020;
originally announced July 2020.
-
An Eddy-Zonal Flow Feedback Model for Propagating Annular Modes
Authors:
Sandro W. Lubis,
Pedram Hassanzadeh
Abstract:
The variability of the zonal-mean large-scale extratropical circulation is often studied using individual modes obtained from empirical orthogonal function (EOF) analyses. The prevailing reduced-order model of the leading EOF (EOF1) of zonal-mean zonal wind, called the annular mode, consists of an eddy-mean flow interaction mechanism that results in a positive feedback of EOF1 onto itself. However…
▽ More
The variability of the zonal-mean large-scale extratropical circulation is often studied using individual modes obtained from empirical orthogonal function (EOF) analyses. The prevailing reduced-order model of the leading EOF (EOF1) of zonal-mean zonal wind, called the annular mode, consists of an eddy-mean flow interaction mechanism that results in a positive feedback of EOF1 onto itself. However, a few studies have pointed out that under some circumstances in observations and GCMs, strong couplings exist between EOF1 and EOF2 at some lag times, resulting in decaying-oscillatory, or propagating, annular modes. Here, we introduce a reduced-order model for coupled EOF1 and EOF2 that accounts for potential cross-EOF eddy-zonal flow feedbacks. Using the analytical solution of this model, we derive conditions for the existence of the propagating regime based on the feedback strengths. Using this model, and idealized GCMs and stochastic prototypes, we show that cross-EOF feedbacks play an important role in controlling the persistence of the annular modes by setting the frequency of the oscillation. We find that stronger cross-EOF feedbacks lead to less persistent annular modes. Applying the coupled-EOF model to the Southern Hemisphere reanalysis data shows the existence of strong cross-EOF feedbacks. The results highlight the importance of considering the coupling of EOFs and cross-EOF feedbacks to fully understand the natural and forced variability of the zonal-mean large-scale circulation.
△ Less
Submitted 16 July, 2020;
originally announced July 2020.
