Based on differences in dynamic response times in the crop production process, a hierarchical decomposition of greenhouse climate management is proposed. To a large extent the proposed decomposition builds on the time-scale decomposition of singularly perturbed systems commonly found in the literature. Main difference with these existing theoretical concepts is that the proposed decomposition is able to deal with rapidly fluctuating deterministic external inputs or disturbances acting on the fast sub-processes. For an example of economic optimal greenhouse climate management during one lettuce production cycle, the decomposition was successfully evaluated in simulations. Using these favourable results, a hierarchical concept for economic optimal greenhouse climate management is derived and discussed in view of application in horticultural practice.