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 1

NN architecture (inputs and outputs) investigated in the paper.

NN architecture	NN inputs		NN outputs
In : out	T	QV	Q1C	Q2	PREC	CLD
36 : 55	18	18	18	18	1	18

T is temperature, QV is atmospheric moisture—vapor mixing ratio, Q1C: the “apparent heat source,” Q2: the “apparent moist sink,” PREC: precipitation rates, and CLD: cloudiness. Numbers in the table show the dimensionality of the corresponding input and output parameters. In : Out stand for NN inputs and outputs and show their corresponding numbers.