We report the results of a series of numerical simulations of convective dynamos, analysis of which points out the important role of small-scale magnetic fields in highly-nonlinear convective dynamos.

In the first part of the talk we show that in our simulations, with the increase of Reynolds number, the magnitude of current helicity increases dramatically, whereas the variation of kinetic helicity is moderate. The competition between the kinetic helicity term and the current helicity term of the alpha coefficient results in an interesting phenomena of the large-scale magnetic fields that resembles the “dynamo-disappear-and-recover” phenomena reported in Hotta et al. (2016). Our simulation and analysis indicate that the role of current helicity first functions to suppress the dynamo, as the convectional alpha-quenching concept states, but then functions to drive the dynamo, instead of quenching it, after a critical Reynolds number is exceeded.

In the second part of the talk we present the properties of the mean-field coefficients that we extract from our dynamo simulations. Our analysis confirms Pouquet et al.(1976)’s formula and proves that current helicity plays an important and possibly dominate role. Unlike what the convectional alpha-quenching formula predicts, in some regions the alpha coefficient is actually larger during the maximum. In the high Reynolds number case, the beta coefficient in some regions can become negative. All these point out the importance of the magnetic terms in the mean-field coefficients.