The capture of system requirements with accuracy and precision remains a challenge for space systems. There is fundamental reliance within current research efforts on natural language (shall statements), which is inherently ambiguous, and thus unable to capture the problem space accurately and precisely. In this article, we suggest an approach to modeling the requirements that leverages forma modeling and system theoretic construct as opposed to reliance on natural language. The proposed approach, specifically, extends structural and behavioral model elements of the Systems Modeling Language (SysML) through definition of a solution space in system-theoretic context. We consider a system to be a transformation of inputs into outputs; and as such model the problem space as a required set of transformation of inputs into outputs. The application of the proposed model-based requirements approach is demonstrated for a notional fire detection, Earth observation satellite FireSat.