Deficiencies in convection trigger functions, used in deep convection parameterizations in General Circulation Models (GCMs), have critical impacts on climate simulations. A novel convection trigger function is developed using the machine learning (ML) classification model XGBoost. The large‐scale environmental information associated with convective events is obtained from the long‐term constrained variational analysis forcing data from the Atmospheric Radiation Measurement (ARM) program at its Southern Great Plains (SGP) and Manaus (MAO) sites representing, respectively, continental mid‐latitude and tropical convection. The ML trigger is separately trained and evaluated per site, and jointly trained and evaluated at both sites as a unified trigger. The performance of the ML trigger is compared with four convective trigger functions commonly used in GCMs: dilute convective available potential energy (CAPE), undilute CAPE, dilute dynamic CAPE (dCAPE), and undilute dCAPE. The ML trigger substantially outperforms the four CAPE‐based triggers in terms of theF1 score metric, widely used to estimate the performance of ML methods. The site‐specific ML trigger functions can achieve, respectively, 91% and 93%F1 scores at SGP and MAO. The unified trigger also has a 91%F1 score, with virtually no degradation from the site‐specific training, suggesting the potential of a global ML trigger function. The ML trigger alleviates a GCM deficiency regarding the overprediction of convection occurrence, offering a promising improvement to the simulation of the diurnal cycle of precipitation. Furthermore, to overcome the black box issue of the ML methods, insights derived from the ML model are discussed, which may be leveraged to improve traditional CAPE‐based triggers.

Tropical cyclone (TC) genesis is a problem of great significance in climate and weather research. Although various environmental conditions necessary for TC genesis have been recognized for a long time, prediction of TC genesis remains a challenge due to complex and stochastic processes involved during TC genesis. Different from traditional statistical and dynamical modeling of TC genesis, in this study, a machine learning framework is developed to determine whether a mesoscale convective system (MCS) would evolve into a tropical cyclone. The machine learning models 1) are built upon a number of essential environmental predictors associated with MCSs/TCs, 2) predict whether MCSs can become TCs at different lead times, and 3) provide information about the relative importance of each predictor, which can be conducive to discovering new aspects of TC genesis. The results indicate that the machine learning classifier, AdaBoost, is able to achieve a 97.2% F1-score accuracy in predicting TC genesis over the entire tropics at a 6-h lead time using a comprehensive set of environmental predictors. A robust performance can still be attained when the lead time is extended to 12, 24, and 48 h, and when this machine learning classifier is separately applied to the North Atlantic Ocean and the western North Pacific Ocean. In contrast, the conventional approach based on the genesis potential index can have no more than an 80% F1-score accuracy. Furthermore, the machine learning classifier suggests that the low-level vorticity and genesis potential index are the most important predictors to TC genesis, which is consistent with previous discoveries.

Accurate short-term precipitation forecast is extremely important for urban flood warning and natural disaster prevention. In this paper, we present an innovative deep learning model named ISA-PredRNN (improved self-attention PredRNN) for precipitation nowcasting based on radar echoes on the basis of the advanced PredRNN-V2. We introduce the self-attention mechanism and the long-term memory state into the model and design a new set of gating mechanisms. To better capture different intensities of precipitation, the loss function with weights was designed. We further train the model using a combination of reverse scheduled sampling and scheduled sampling to learn the long-term dynamics from the radar echo sequences. Experimental results show that the new model (ISA-PredRNN) can effectively extract the spatiotemporal features of radar echo maps and obtain radar echo prediction results with a small gap from the ground truths. From the comparison with the other six models, the new ISA-PredRNN model has the most accurate prediction results with a critical success index (CSI) of 0.7001, 0.5812 and 0.3052 under the radar echo thresholds of 10 dBZ, 20 dBZ and 30 dBZ, respectively.

This paper describes the atmospheric component of the US Department of Energy's Energy Exascale Earth System Model (E3SM) version 3. Significant updates have been made to the atmospheric physics compared to earlier versions. Specifically, interactive gas chemistry has been implemented, along with improved representations of aerosols and dust emissions. A new stratiform cloud microphysics scheme more physically treats ice processes and aerosol‐cloud interactions. The deep convection parameterization has been largely improved with sophisticated microphysics for convective clouds, making model convection sensitive to large‐scale dynamics, and incorporating the dynamical and physical effects of organized mesoscale convection. Improvements in aerosol wet removal processes and parameter re‐tuning of key aerosol and cloud processes have improved model aerosol radiative forcing. The model's vertical resolution has increased from 72 to 80 layers with the extra eight layers added in the lower stratosphere to better simulate the Quasi‐Biennial Oscillation. These improvements have enhanced E3SM's capability to couple aerosol, chemistry, and biogeochemistry and reduced some long‐standing biases in simulating tropical variability. Compared to its predecessors, the model shows a much stronger signal for the Madden‐Julian Oscillation, Kelvin waves, mixed Rossby‐gravity waves, and eastward inertia‐gravity waves. Aerosol radiative forcing has been considerably reduced and is now better aligned with community best estimates, leading to significantly improved skill in simulating historical temperature records. Its simulated mean‐state climate is largely comparable to E3SMv2, but with some notable degradation in shortwave cloud radiative effect, precipitable water, and surface wind stress, which will be addressed in future updates.

Traditional trial-and-error tuning of uncertain parameters in global atmospheric general circulation models (GCMs) is time consuming and subjective. This study explores the feasibility of automatic optimization of GCM parameters for fast physics by using short-term hindcasts. An automatic workflow is described and applied to the Community Atmospheric Model (CAM5) to optimize several parameters in its cloud and convective parameterizations. We show that the auto-optimization leads to 10 % reduction of the overall bias in CAM5, which is already a well-calibrated model, based on a predefined metric that includes precipitation, temperature, humidity, and longwave/shortwave cloud forcing. The computational cost of the entire optimization procedure is about equivalent to a single 12-year atmospheric model simulation. The tuning reduces the large underestimation in the CAM5 longwave cloud forcing by decreasing the threshold relative humidity and the sedimentation velocity of ice crystals in the cloud schemes; it reduces the overestimation of precipitation by increasing the adjustment time in the convection scheme. The physical processes behind the tuned model performance for each targeted field are discussed. Limitations of the automatic tuning are described, including the slight deterioration in some targeted fields that reflect the structural errors of the model. It is pointed out that automatic tuning can be a viable supplement to process-oriented model evaluations and improvement.

Wildfires have shown increasing trends in both frequency and severity across the contiguous United States (CONUS). However, process-based fire models have difficulties in accurately simulating the burned area over the CONUS due to a simplification of the physical process and cannot capture the interplay among fire, ignition, climate, and human activities. The deficiency of burned area simulation deteriorates the description of fire impact on energy balance, water budget, and carbon fluxes in the Earth system models (ESMs). Alternatively, fire models based on machine learning (ML), which capture statistical relationships between the burned area and environmental factors, have shown promising burned area predictions and corresponding fire impact simulation. We develop a hybrid framework (ELM2.1-XGBFire1.0) that integrates an eXtreme Gradient Boosting (XGBoost) wildfire model with the Energy Exascale Earth System Model (E3SM) land model (ELM) version 2.1. A Fortran–C–Python deep learning bridge is adapted to support online communication between ELM and the ML fire model. Specifically, the burned area predicted by the ML-based wildfire model is directly passed to ELM to adjust the carbon pool and vegetation dynamics after disturbance, which are then used as predictors in the ML-based fire model in the next time step. Evaluated against the historical burned area from Global Fire Emissions Database 5 from 2001–2019, the ELM2.1-XGBFire1.0 outperforms process-based fire models in terms of spatial distribution and seasonal variations. The ELM2.1-XGBFire1.0 has proven to be a new tool for studying vegetation–fire interactions and, more importantly, enables seamless exploration of climate–fire feedback, working as an active component of E3SM.