Software ecosystems is one promising strategy for organizations to find new market segments, new innovative value propositions creating new value streams. However, understanding internal and external actors, resources and relationships that could be leveraged in a SECO is critical for their strategic decisions. The consequence of mistakes may be costly failures that can force an organization to move out of a market. This paper describes a systematic mapping study that targets description of software ecosystems. Our conjecture is that adequate description support leads to modeling, which will improve information and in turn strategic decisions. The survey searches existing literature for description techniques and their application for comprehensive description. The study identifies and maps 63 primary studies out of 937 candidates according to their degree of modeling support and several other important aspects for SECO description. The analysis indicates that no approach fully supports comprehensive SECO descriptions, supporting domain specific and view specific modeling of ecosystem concerns. The analysis is used to highlight areas for a future research agenda.