In FY2019, NCAR scientists further investigated the mechanisms by which short-timescale perturbations to atmospheric processes can affect interannual variability such as the El Niño-Southern Oscillation (ENSO). To this end a coupled simulation of NCAR's Community Climate System Model was compared to a simulation in which the model's atmospheric diabatic tendencies are perturbed at each time step using a Stochastically Perturbed Parameterized Tendencies scheme.

The simulation with stochastic parameterization compares better with 20th-century reanalysis in having lower inter-annual sea surface temperature (SST) variability and more irregular transitions between El Niño and La Niña states (Fig). The broadening of the SST spectrum is consistent with a reduced decorrelation time scale of the ENSO eigenmode from 17 to 11 months, which is closer to 8 months as obtained in 20th-century reanalysis. It was shown that for the simple model of a linear damped oscillator, such a noise-induced stabilization and broadening of the spectrum is obtained when its frequency parameter is perturbed (i.e., a randomization of phases).

These results can be used to inform next generation physical parameterization development. NCAR scientists will continue study of other areas where fast-slow interactions play a fundamental role and stochastic parameterization can add value to weather predictions and climate projections.

These findings were published in the Journal of Climate.

Figure: Power spectrum of the Niño 3.4-index, defined as the monthly SST anomaly averaged over 5S-5N and 170E-120W, for HadISST2 observations (black), for control experiment (blue) and experiment with stochastically perturbed parameterization tendencies SPPT (red). Top axis indicates period in years and bottom axis frequency in cycles per month. The shading denotes the spectral range obtained by sampling realizations from linear inverse models fitted to CNTL, SPPT and HadISST2.