Using Ensemble of Neural Networks to Learn Stochastic Convection Parameterizations for Climate and Numerical Weather Prediction Models from Data Simulated by a Cloud Resolving Model

<table>Development design of a NN convection parameterization. <i>T, Q, </i>and so forth refer to selected CRM-simulated fields used as inputs for NN parameterization; Prec., Tendencies, and so forth refer to selected CRM simulated variables used as NN outputs. The dotted and dashed lines indicate that observational and high-resolution simulated data can be added to the training data set if necessary.</table>

Advances in Artificial Neural Systems

fig1

Figure 1

Figure 1: Using Ensemble of Neural Networks to Learn Stochastic Convection Parameterizations for Climate and Numerical Weather Prediction Models from Data Simulated by a Cloud Resolving Model 

Figure 1 | Using Ensemble of Neural Networks to Learn Stochastic Convection Parameterizations for Climate and Numerical Weather Prediction Models from Data Simulated by a Cloud Resolving Model 