Modern environmental problems pose unique management challenges as they are usually interdependent in myriad, complex ways. Climate change is the ultimate example of a problem that forces environmental managers to confront highly interdependent challenges, such as invasive species, rising temperatures, and habitat loss. Interdependencies abound: for example, the issue of warming winter temperatures exacerbates the issue of invasive species, and a high prevalence of invasive species contributes to the issue of habitat loss. Ideally, stakeholders should account for these issue interdependencies by managing connected issues. Such activities close "integrative gaps," which refer to instances in which interdependent issues are managed independently. By closing integrative gaps, actors align management activities with underlying ecological processes. We focus on climate change adaptation governance in Ohio, USA, as a model study system and evaluate conditions that enable integrative gap closure through analysis of a network of adaptation actors and issues. Our findings show that actors are more likely to close integrative gaps between issue pairs that are highly biophysically interdependent, receive higher collective levels of public attention, and have garnered higher levels of progress. We also find that regional-scoped, specialized, and non-profit actors are most likely to manage for environmental interdependencies. We discuss how these findings advance theoretical understanding of institutional fitness and resilience in socialecological systems by revealing how actors navigate highly interdependent environmental governance settings.

Understanding how stakeholders choose to participate in different policy forums is central to research on complex, polycentric governance systems. In this paper, we draw upon the Ecology of Games Theory (EGT) to develop theoretical expectations about how four incentive structures may guide how actors navigate the world of policy forums. We test these expectations using unique data on a three-mode network of actors, forums, and issues related to climate change adaption in the state of Ohio, in the U.S. Midwest. Results of an exponential random graph model suggest that multilevel closure structures, which are a function of transaction costs and direct benefits, guide actors’ forum participation in ways that can either reinforce sub-optimal, ineffective governance arrangements, or conversely, encourage opportunities for innovation, increase diversity in representation, and facilitate policy learning. From a methodological standpoint, our research highlights the benefits of examining complex governance systems through the more precise approach allowed by three-mode network analysis, which has not been frequently used in research on polycentric governance systems up to this point.