With the recent advances in data science, machine learning has been increasingly applied to convection and cloud parameterizations in global climate models (GCMs). This study extends the work of Han et al. (2020,
We explore the potential of feed‐forward deep neural networks (DNNs) for emulating cloud superparameterization in realistic geography, using offline fits to data from the superparameterized community atmospheric model. To identify the network architecture of greatest skill, we formally optimize hyperparameters using ∼250 trials. Our DNN explains over 70% of the temporal variance at the 15‐min sampling scale throughout the mid‐to‐upper troposphere. Autocorrelation timescale analysis compared against DNN skill suggests the less good fit in the tropical, marine boundary layer is driven by neural network difficulty emulating fast, stochastic signals in convection. However, spectral analysis in the temporal domain indicates skillful emulation of signals on diurnal to synoptic scales. A closer look at the diurnal cycle reveals correct emulation of land‐sea contrasts and vertical structure in the heating and moistening fields, but some distortion of precipitation. Sensitivity tests targeting precipitation skill reveal complementary effects of adding positive constraints versus hyperparameter tuning, motivating the use of both in the future. A first attempt to force an offline land model with DNN emulated atmospheric fields produces reassuring results further supporting neural network emulation viability in real‐geography settings. Overall, the fit skill is competitive with recent attempts by sophisticated Residual and Convolutional Neural Network architectures trained on added information, including memory of past states. Our results confirm the parameterizability of superparameterized convection with continents through machine learning and we highlight the advantages of casting this problem locally in space and time for accurate emulation and hopefully quick implementation of hybrid climate models.
more » « less- NSF-PAR ID:
- 10360366
- Publisher / Repository:
- DOI PREFIX: 10.1029
- Date Published:
- Journal Name:
- Journal of Advances in Modeling Earth Systems
- Volume:
- 13
- Issue:
- 5
- ISSN:
- 1942-2466
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
Abstract https://doi.org/10.1029/2020MS002076 ) and uses an ensemble of 32‐layer deep convolutional residual neural networks, referred to as ResCu‐en, to emulate convection and cloud processes simulated by a superparameterized GCM, SPCAM. ResCu‐en predicts GCM grid‐scale temperature and moisture tendencies, and cloud liquid and ice water contents from moist physics processes. The surface rainfall is derived from the column‐integrated moisture tendency. The prediction uncertainty inherent in deep learning algorithms in emulating the moist physics is reduced by ensemble averaging. Results in 1‐year independent offline validation show that ResCu‐en has high prediction accuracy for all output variables, both in the current climate and in a warmer climate with +4K sea surface temperature. The analysis of different neural net configurations shows that the success to generalize in a warmer climate is attributed to convective memory and the 1‐dimensional convolution layers incorporated into ResCu‐en. We further implement a member of ResCu‐en into CAM5 with real world geography and run the neural‐network‐enabled CAM5 (NCAM) for 5 years without encountering any numerical integration instability. The simulation generally captures the global distribution of the mean precipitation, with a better simulation of precipitation intensity and diurnal cycle. However, there are large biases in temperature and moisture in high latitudes. These results highlight the importance of convective memory and demonstrate the potential for machine learning to enhance climate modeling. -
null (Ed.)Abstract Neural networks are a promising technique for parameterizing subgrid-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 midtropospheric 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 superparameterized 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 −1 . In contrast, standing unstable modes do not cause catastrophic instability. Using these tools, differences between the SPCAM-trained versus GCRM-trained neural networks are analyzed, and strategies to incrementally improve both of their coupled online performance unveiled.more » « less
-
more » « less
Abstract Attempts to use machine learning to develop atmospheric parameterizations have mainly focused on subgrid effects on temperature and moisture, but subgrid momentum transport is also important in simulations of the atmospheric circulation. Here, we use neural networks to develop a subgrid momentum transport parameterization that learns from coarse‐grained output of a high‐resolution atmospheric simulation in an idealized aquaplanet domain. We show that substantial subgrid momentum transport occurs due to convection. The neural‐network parameterization has skill in predicting momentum fluxes associated with convection, although its skill for subgrid momentum fluxes is lower compared to subgrid energy and moisture fluxes. The parameterization conserves momentum, and when implemented in the same atmospheric model at coarse resolution it leads to stable simulations and tends to reduce wind biases, although it over‐corrects for one configuration tested. Overall, our results show that it is challenging to predict subgrid momentum fluxes and that machine‐learning momentum parameterization gives promising results.
-
Abstract. This study examines the diurnal variation in precipitation over Hainan Island in the South China Sea using gauge observations from 1951 to 2012 and Climate Prediction Center MORPHing technique (CMORPH) satellite estimates from 2006 to 2015, as well as numerical simulations. The simulations are the first to use climatological mean initial and lateral boundary conditions to study the dynamic and thermodynamic processes (and the impacts of land–sea breeze circulations) that control the rainfall distribution and climatology. Precipitation is most significant from April to October and exhibits a strong diurnal cycle resulting from land–sea breeze circulations. More than 60% of the total annual precipitation over the island is attributable to the diurnal cycle with a significant monthly variability. The CMORPH and gauge datasets agree well, except that the CMORPH data underestimate precipitation and have a 1 h peak delay. The diurnal cycle of the rainfall and the related land– sea breeze circulations during May and June were well captured by convection-permitting numerical simulations with the Weather Research and Forecasting (WRF) model, which were initiated from a 10-year average ERA-Interim reanalysis. The simulations have a slight overestimation of rainfall amounts and a 1 h delay in peak rainfall time. The diurnal cycle of precipitation is driven by the occurrence of moist convection around noontime owing to low-level convergence associated with the sea-breeze circulations. The precipitation intensifies rapidly thereafter and peaks in the afternoon with the collisions of sea-breeze fronts from different sides of the island. Cold pools of the convective storms contribute to the inland propagation of the sea breeze. Generally, precipitation dissipates quickly in the evening due to the cooling and stabilization of the lower troposphere and decrease of boundary layer moisture. Interestingly, the rather high island orography is not a dominant factor in the diurnal variation in precipitation over the island.more » « less
-
more » « less
Abstract A combination of cloud-permitting model (CPM) simulations, satellite, and reanalysis data are used to test whether the diurnal cycle in surface temperature has a significant impact on the intensity of deep convection as measured by high-percentile updraft velocities, lightning, and CAPE. The land–ocean contrast in lightning activity shows that convective intensity varies between land and ocean independently from convective quantity. Thus, a mechanism that explains the land–ocean contrast must be able to do so even after controlling for precipitation variations. Motivated by the land–ocean contrast, we use idealized CPM simulations to test the impact of the diurnal cycle on high-percentile updrafts. In simulations, updrafts are somewhat enhanced due to large-scale precipitation enhancement by the diurnal cycle. To control for large-scale precipitation, we use statistical sampling techniques. After controlling for precipitation enhancement, the diurnal cycle does not affect convective intensities. To explain why sampled updrafts are not enhanced, we note that CAPE is also not increased, likely due to boundary layer quasi equilibrium (BLQE) occurring over our land area. Analysis of BLQE in terms of net positive and negative mass flux finds that boundary layer entrainment, and even more importantly downdrafts, account for most of the moist static energy (MSE) sink that is balancing surface fluxes. Using ERA-Interim data, we also find qualitative evidence for BLQE over land in the real world, as high percentiles of CAPE are not greater over land than over ocean.
