Tom Beucler

Bio: Tom Beucler is an academic researcher from University of California, Irvine. The author has contributed to research in topics: Physics & Artificial neural network. The author has an hindex of 8, co-authored 24 publications receiving 274 citations. Previous affiliations of Tom Beucler include Columbia University & Massachusetts Institute of Technology.

Enforcing Analytic Constraints in Neural Networks Emulating Physical Systems.

Abstract: Neural networks can emulate nonlinear physical systems with high accuracy, yet they may produce physically inconsistent results when violating fundamental constraints. Here, we introduce a systematic way of enforcing nonlinear analytic constraints in neural networks via constraints in the architecture or the loss function. Applied to convective processes for climate modeling, architectural constraints enforce conservation laws to within machine precision without degrading performance. Enforcing constraints also reduces errors in the subsets of the outputs most impacted by the constraints.

Achieving Conservation of Energy in Neural Network Emulators for Climate Modeling

Abstract: Artificial neural-networks have the potential to emulate cloud processes with higher accuracy than the semi-empirical emulators currently used in climate models. However, neural-network models do not intrinsically conserve energy and mass, which is an obstacle to using them for long-term climate predictions. Here, we propose two methods to enforce linear conservation laws in neural-network emulators of physical models: Constraining (1) the loss function or (2) the architecture of the network itself. Applied to the emulation of explicitly-resolved cloud processes in a prototype multi-scale climate model, we show that architecture constraints can enforce conservation laws to satisfactory numerical precision, while all constraints help the neural-network better generalize to conditions outside of its training set, such as global warming.

Interpreting and Stabilizing Machine-learning Parametrizations of Convection

Abstract: Neural networks are a promising technique for parameterizing sub-grid-scale physics (e.g. moist atmospheric convection) in coarse-resolution climate models, but their lack of interpretability and reliability prevents widespread adoption. For instance, it is not fully understood why neural network parameterizations often cause dramatic instability when coupled to atmospheric fluid dynamics. This paper introduces tools for interpreting their behavior that are customized to the parameterization task. First, we assess the nonlinear sensitivity of a neural network to lower-tropospheric stability and the mid-tropospheric moisture, two widely-studied controls of moist convection. Second, we couple the linearized response functions of these neural networks to simplified gravity-wave dynamics, and analytically diagnose the corresponding phase speeds, growth rates, wavelengths, and spatial structures. To demonstrate their versatility, these techniques are tested on two sets of neural networks, one trained with a super-parametrized version of the Community Atmosphere Model (SPCAM) and the second with a near-global cloud-resolving model (GCRM). Even though the SPCAM simulation has a warmer climate than the cloud-resolving model, both neural networks predict stronger heating/drying in moist and unstable environments, which is consistent with observations. Moreover, the spectral analysis can predict that instability occurs when GCMs are coupled to networks that support gravity waves that are unstable and have phase speeds larger than 5 m/s. In contrast, standing unstable modes do not cause catastrophic instability. Using these tools, differences between the SPCAM- vs. GCRM- trained neural networks are analyzed, and strategies to incrementally improve both of their coupled online performance unveiled.

Moisture-radiative cooling instability

Abstract: Radiative-convective equilibrium (RCE) - the statistical equilibrium state of the atmosphere where convection and radiation interact in the absence of lateral transport - is widely used as a basic-state model of the tropical atmosphere. The possibility that RCE may be unstable to development of large-scale circulation has been raised by recent modeling, theoretical, and observational studies, and could have profound consequences for our understanding of tropical meteorology and climate. Here, we study the interaction between moisture and radiative cooling as a contributor to instability of RCE. We focus on whether the total atmospheric radiative cooling decreases with column water vapor; this condition, which we call moisture-radiative cooling instability (MRCI), provides the potential for unstable growth of moist or dry perturbations. Analytic solutions to the gray-gas radiative transfer equations show that MRCI is satisfied when the total column optical depth - linked to column water vapor - exceeds a critical threshold. Both the threshold and the growth rate of the instability depend strongly on the shape of the water vapor perturbation. Calculations with a realistic radiative transfer model confirm the existence of MRCI for typical tropical values of column water vapor, but show even stronger dependence on the vertical structure of water vapor perturbation. Finally, we analyze the sensitivity of atmospheric radiative cooling to variability in column water vapor in observed tropical soundings. We find that clear-sky MRCI is satisfied across a range of locations and seasons in the real tropical atmosphere, with a partial growth rate of ∼1 month−1 This article is protected by copyright. All rights reserved.

Interpreting and Stabilizing Machine-learning Parametrizations of Convection

Abstract: Neural networks are a promising technique for parameterizing sub-grid-scale physics (e.g. moist atmospheric convection) in coarse-resolution climate models, but their lack of interpretability and reliability prevents widespread adoption. For instance, it is not fully understood why neural network parameterizations often cause dramatic instability when coupled to atmospheric fluid dynamics. This paper introduces tools for interpreting their behavior that are customized to the parameterization task. First, we assess the nonlinear sensitivity of a neural network to lower-tropospheric stability and the mid-tropospheric moisture, two widely-studied controls of moist convection. Second, we couple the linearized response functions of these neural networks to simplified gravity-wave dynamics, and analytically diagnose the corresponding phase speeds, growth rates, wavelengths, and spatial structures. To demonstrate their versatility, these techniques are tested on two sets of neural networks, one trained with a super-parametrized version of the Community Atmosphere Model (SPCAM) and the second with a near-global cloud-resolving model (GCRM). Even though the SPCAM simulation has a warmer climate than the cloud-resolving model, both neural networks predict stronger heating/drying in moist and unstable environments, which is consistent with observations. Moreover, the spectral analysis can predict that instability occurs when GCMs are coupled to networks that support gravity waves that are unstable and have phase speeds larger than 5 m/s. In contrast, standing unstable modes do not cause catastrophic instability. Using these tools, differences between the SPCAM- vs. GCRM- trained neural networks are analyzed, and strategies to incrementally improve both of their coupled online performance unveiled.

Integrating Physics-Based Modeling with Machine Learning: A Survey

Abstract: In this manuscript, we provide a structured and comprehensive overview of techniques to integrate machine learning with physics-based modeling. First, we provide a summary of application areas for which these approaches have been applied. Then, we describe classes of methodologies used to construct physics-guided machine learning models and hybrid physics-machine learning frameworks from a machine learning standpoint. With this foundation, we then provide a systematic organization of these existing techniques and discuss ideas for future research.

Coherent Intraseasonal Oscillations of Ocean and Atmosphere During the Asian Summer Monsoon

Abstract: The space-time evolution of the ocean and atmosphere associated with 1998-2000 monsoon intraseasonal oscillations (ISO) in the Indian Ocean and west Pacific is studied using validated sea surface temperature (SST) and surface wind speed from the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager, and satellite outgoing longwave radiation. Monsoon ISO consist of alternating episodes of active and suppressed atmospheric convection moving northward in the eastern Indian Ocean and the South China Sea. Negative/positive SST anomalies generated by fluctuations of net heat flux at the ocean surface move northward following regions of active/suppressed convection. Such coherent evolution of SST, surface heat flux and convection suggests that air-sea interaction might be important in monsoon ISO.

Enforcing Analytic Constraints in Neural Networks Emulating Physical Systems.

Abstract: Neural networks can emulate nonlinear physical systems with high accuracy, yet they may produce physically inconsistent results when violating fundamental constraints. Here, we introduce a systematic way of enforcing nonlinear analytic constraints in neural networks via constraints in the architecture or the loss function. Applied to convective processes for climate modeling, architectural constraints enforce conservation laws to within machine precision without degrading performance. Enforcing constraints also reduces errors in the subsets of the outputs most impacted by the constraints.

What Role Does Hydrological Science Play in the Age of Machine Learning

Abstract: We suggest that there is a potential danger to the hydrological sciences community in not recognizing how transformative machine learning will be for the future of hydrological modeling. Given the recent success of machine learning applied to modeling problems, it is unclear what the role of hydrological theory might be in the future. We suggest that a central challenge in hydrology right now should be to clearly delineate where and when hydrological theory adds value to prediction systems. Lessons learned from the history of hydrological modeling motivate several clear next steps toward integrating machine learning into hydrological modeling workflows.