Stochastic linear models of nonlinear geosystems

When nonlinearities are strong enough, they can often be treated in terms of linear stochastic differential equations. We discuss the conditions under which this can be done, and illustrate the linear approximation with several examples having varying success. In particular, we consider the low-dimensional chaotic Lorenz system, a linear process driven by the chaotic Lorenz system, and the real-world application to tropical sea surface temperatures.