Jin-Song von Storch

In my research, I study the Earth's climate as a high-dimensional forced dissipative system.  In doing so, I take two approaches. One is process-oriented. I focus on the ocean and study the processes that control and maintain the ocean's general circulation. From an energetics perspective, these are (a) processes that provide mechanical energy for diapycnal mixing in the ocean interior and thus support the upwelling required to close an overturning circulation, and (b) processes that dissipate mechanical energy and thus can counteract the constant supply of energy from wind and buoyancy forcing at the sea surface.  I believe that understanding these processes is essential for understanding how the ocean works. The other approach is systems-oriented. A climate model is a set of discretized differential equations that have the same general structure as, say, a many-particle system considered in statistical physics. Climate variables exhibit random fluctuations similar to those observed in phenomena such as Brownian motion in statistical physics. These similarities suggest that the methods and concepts developed for statistical physics may also be applicable to climate and vice versa. The goal is to produce probabilistic climate predictions that describe changes in climate statistics caused by changes in an external forcing. Here we use probability as a measure of the inherent randomness of the climate system (most evident when considering internal climate variability) rather than as a measure of uncertainties arising from the lack of knowledge (a standard argument used to justify probabilistic reasoning).